[CI] Add multi-nodes EPLB configs of DeepSeek-R1-W8A8 & Qwen3-235B-W8A8 (#4144)

### What this PR does / why we need it?
add DeepSeek-R1-W8A8 and Qwen3-235B-W8A8 configs in multi-nodes and EPLB
scenario

### Does this PR introduce _any_ user-facing change?
no

- vLLM version: v0.11.0
- vLLM main:
83f478bb19

---------

Signed-off-by: 白永斌 <baiyongbin3@h-partners.com>
Co-authored-by: 白永斌 <baiyongbin3@h-partners.com>

.github/workflows/vllm_ascend_test_nightly_a3.yaml

@@ -134,6 +134,12 @@ jobs:
           - name: multi-node-dpsk3.2-exp-2node
             config_file_path: DeepSeek-V3_2-Exp-bf16.yaml
             size: 2
+          - name: multi-node-deepseek-r1-w8a8-eplb
+            config_file_path: DeepSeek-R1-W8A8-EPLB.yaml
+            size: 4
+          - name: multi-node-qwenw8a8-2node-eplb
+            config_file_path: Qwen3-235B-W8A8-EPLB.yaml
+            size: 2
     uses: ./.github/workflows/_e2e_nightly_multi_node.yaml
     with:
       soc_version: a3

tests/e2e/nightly/multi_node/config/models/DeepSeek-R1-W8A8-EPLB.yaml

@@ -0,0 +1,165 @@
+test_name: "test DeepSeek-R1-W8A8 disaggregated_prefill"
+model: "vllm-ascend/DeepSeek-R1-0528-W8A8"
+num_nodes: 4
+npu_per_node: 16
+env_common:
+  VLLM_USE_MODELSCOPE: true
+  HCCL_BUFFSIZE: 1024
+  SERVER_PORT: 8080
+  OMP_PROC_BIND: false
+  OMP_NUM_THREADS: 10
+  PYTORCH_NPU_ALLOC_CONF: expandable_segments:True
+  HCCL_DETERMINISTIC: True
+  TASK_QUEUE_ENABLE: 1
+  HCCL_OP_RETRY_ENABLE: "L0:0, L1:0, L2:0"
+  DYNAMIC_EPLB: true
+
+disaggregated_prefill:
+  enabled: true
+  prefiller_host_index: [0, 1]
+  decoder_host_index: [2]
+  ranktable_gen_path: "examples/disaggregated_prefill_v1/gen_ranktable.py"
+  ranktable_path: "/tmp/ranktable.json"
+
+deployment:
+  -
+    server_cmd: >
+      vllm serve vllm-ascend/DeepSeek-R1-0528-W8A8
+          --host 0.0.0.0
+          --port $SERVER_PORT
+          --data-parallel-size 2
+          --data-parallel-size-local 2
+          --tensor-parallel-size 8
+          --enforce-eager
+          --enable-expert-parallel
+          --seed 1024
+          --quantization ascend
+          --max-num-seqs 4
+          --max-model-len 36864
+          --max-num-batched-tokens 16384
+          --trust-remote-code
+          --gpu-memory-utilization 0.9
+          --speculative-config '{"num_speculative_tokens": 1, "method":"deepseek_mtp"}'
+          --kv-transfer-config
+          '{"kv_connector": "LLMDataDistCMgrConnector",
+          "kv_buffer_device": "npu",
+          "kv_role": "kv_producer",
+          "kv_parallel_size": 1,
+          "kv_port": "20001",
+          "engine_id": "0",
+          "kv_connector_module_path": "vllm_ascend.distributed.llmdatadist_c_mgr_connector"
+          }'
+          --additional-config
+          '{"ascend_scheduler_config":{"enabled":false},"torchair_graph_config":{"enabled":false,"enable_multistream_shared_expert":false},"enable_prefill_optimizations":true,"enable_weight_nz_layout":true,"dynamic_eplb":true,"num_iterations_eplb_update":2048,"num_wait_worker_iterations":200}'
+
+  -
+    server_cmd: >
+      vllm serve vllm-ascend/DeepSeek-R1-0528-W8A8
+          --host 0.0.0.0
+          --port $SERVER_PORT
+          --data-parallel-size 2
+          --data-parallel-size-local 2
+          --tensor-parallel-size 8
+          --enforce-eager
+          --enable-expert-parallel
+          --seed 1024
+          --quantization ascend
+          --max-num-seqs 4
+          --max-model-len 36864
+          --max-num-batched-tokens 16384
+          --trust-remote-code
+          --gpu-memory-utilization 0.9
+          --speculative-config '{"num_speculative_tokens": 1, "method":"deepseek_mtp"}'
+          --kv-transfer-config
+          '{"kv_connector": "LLMDataDistCMgrConnector",
+          "kv_buffer_device": "npu",
+          "kv_role": "kv_producer",
+          "kv_parallel_size": 1,
+          "kv_port": "20001",
+          "engine_id": "1",
+          "kv_connector_module_path": "vllm_ascend.distributed.llmdatadist_c_mgr_connector"
+          }'
+          --additional-config
+          '{"ascend_scheduler_config":{"enabled":false},"torchair_graph_config":{"enabled":false,"enable_multistream_shared_expert":false},"enable_prefill_optimizations":true,"enable_weight_nz_layout":true,"dynamic_eplb":true,"num_iterations_eplb_update":2048,"num_wait_worker_iterations":200}'
+  -
+    server_cmd: >
+      vllm serve vllm-ascend/DeepSeek-R1-0528-W8A8
+        --host 0.0.0.0
+        --port $SERVER_PORT
+        --data-parallel-size 32
+        --data-parallel-size-local 16
+        --data-parallel-start-rank 0
+        --data-parallel-address $LOCAL_IP
+        --data-parallel-rpc-port 13389
+        --tensor-parallel-size 1
+        --enable-expert-parallel
+        --seed 1024
+        --quantization ascend
+        --max-num-seqs 28
+        --max-model-len 36864
+        --max-num-batched-tokens 256
+        --trust-remote-code
+        --gpu-memory-utilization 0.9
+        --speculative-config '{"num_speculative_tokens": 1, "method":"deepseek_mtp"}'
+        --kv-transfer-config
+        '{"kv_connector": "LLMDataDistCMgrConnector",
+        "kv_buffer_device": "npu",
+        "kv_role": "kv_consumer",
+        "kv_parallel_size": 1,
+        "kv_port": "20001",
+        "engine_id": "2",
+        "kv_connector_module_path": "vllm_ascend.distributed.llmdatadist_c_mgr_connector"
+        }'
+        --additional-config
+        '{"ascend_scheduler_config":{"enabled":false},"torchair_graph_config":{"enabled":true,"enable_multistream_mla":true,"graph_batch_sizes":[28],"use_cached_graph":true,"enable_super_kernel":false},"multistream_overlap_shared_expert":true,"dynamic_eplb":true,"num_iterations_eplb_update":2048,"num_wait_worker_iterations":200}'
+  -
+    server_cmd: >
+      vllm serve vllm-ascend/DeepSeek-R1-0528-W8A8
+        --headless
+        --data-parallel-size 32
+        --data-parallel-size-local 16
+        --data-parallel-start-rank 16
+        --data-parallel-address $MASTER_IP
+        --data-parallel-rpc-port 13389
+        --tensor-parallel-size 1
+        --enable-expert-parallel
+        --seed 1024
+        --quantization ascend
+        --max-num-seqs 28
+        --max-model-len 36864
+        --max-num-batched-tokens 256
+        --trust-remote-code
+        --gpu-memory-utilization 0.9
+        --speculative-config '{"num_speculative_tokens": 1, "method":"deepseek_mtp"}'
+        --kv-transfer-config
+        '{"kv_connector": "LLMDataDistCMgrConnector",
+        "kv_buffer_device": "npu",
+        "kv_role": "kv_consumer",
+        "kv_parallel_size": 1,
+        "kv_port": "20001",
+        "engine_id": "2",
+        "kv_connector_module_path": "vllm_ascend.distributed.llmdatadist_c_mgr_connector"
+        }'
+        --additional-config
+        '{"ascend_scheduler_config":{"enabled":false},"torchair_graph_config":{"enabled":true,"enable_multistream_mla":true,"graph_batch_sizes":[28],"use_cached_graph":true,"enable_super_kernel":false},"multistream_overlap_shared_expert":true,"dynamic_eplb":true,"num_iterations_eplb_update":2048,"num_wait_worker_iterations":200}'
+benchmarks:
+  perf:
+    case_type: performance
+    dataset_path: vllm-ascend/GSM8K-in3500-bs2800
+    request_conf: vllm_api_stream_chat
+    dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_str_perf
+    num_prompts: 2800
+    max_out_len: 1500
+    batch_size: 700
+    request_rate: 11.2
+    baseline: 1
+    threshold: 0.97
+  acc:
+    case_type: accuracy
+    dataset_path: vllm-ascend/gsm8k
+    request_conf: vllm_api_general_chat
+    dataset_conf: gsm8k/gsm8k_gen_0_shot_cot_chat_prompt
+    max_out_len: 32768
+    batch_size: 512
+    baseline: 95
+    threshold: 5

tests/e2e/nightly/multi_node/config/models/Qwen3-235B-W8A8-EPLB.yaml

@@ -0,0 +1,91 @@
+test_name: "test Qwen3-235B-A22B-W8A8 disaggregated_prefill"
+model: "vllm-ascend/Qwen3-235B-A22B-W8A8"
+num_nodes: 2
+npu_per_node: 16
+env_common:
+  VLLM_USE_MODELSCOPE: true
+  OMP_PROC_BIND: false
+  OMP_NUM_THREADS: 100
+  HCCL_BUFFSIZE: 1024
+  SERVER_PORT: 8080
+  DYNAMIC_EPLB: true
+disaggregated_prefill:
+  enabled: true
+  prefiller_host_index: [0]
+  decoder_host_index: [1]
+
+deployment:
+  -
+    server_cmd: >
+        vllm serve "vllm-ascend/Qwen3-235B-A22B-W8A8"
+        --host 0.0.0.0
+        --port $SERVER_PORT
+        --data-parallel-size 2
+        --data-parallel-size-local 2
+        --tensor-parallel-size 8
+        --seed 1024
+        --enable-expert-parallel
+        --max-num-seqs 16
+        --max-model-len 8192
+        --max-num-batched-tokens 8192
+        --quantization ascend
+        --trust-remote-code
+        --no-enable-prefix-caching
+        --gpu-memory-utilization 0.9
+        --kv-transfer-config
+        '{"kv_connector": "MooncakeConnector",
+        "kv_role": "kv_producer",
+        "kv_port": "30000",
+        "engine_id": "0",
+        "kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
+        "kv_connector_extra_config": {
+                  "prefill": {
+                          "dp_size": 2,
+                          "tp_size": 8
+                  },
+                  "decode": {
+                          "dp_size": 2,
+                          "tp_size": 8
+                  }
+            }
+        }'
+        --additional-config
+        '{"dynamic_eplb":true,"num_iterations_eplb_update":2048,"num_wait_worker_iterations":200}'
+
+  -
+    server_cmd: >
+        vllm serve "vllm-ascend/Qwen3-235B-A22B-W8A8"
+        --host 0.0.0.0
+        --port $SERVER_PORT
+        --data-parallel-size 2
+        --data-parallel-size-local 2
+        --tensor-parallel-size 8
+        --seed 1024
+        --quantization ascend
+        --max-num-seqs 16
+        --max-model-len 8192
+        --max-num-batched-tokens 8192
+        --enable-expert-parallel
+        --trust-remote-code
+        --no-enable-prefix-caching
+        --gpu-memory-utilization 0.9
+        --kv-transfer-config
+        '{"kv_connector": "MooncakeConnector",
+        "kv_role": "kv_consumer",
+        "kv_port": "30200",
+        "engine_id": "1",
+        "kv_connector_module_path": "vllm_ascend.distributed.mooncake_connector",
+        "kv_connector_extra_config": {
+                  "prefill": {
+                          "dp_size": 2,
+                          "tp_size": 8
+                  },
+                  "decode": {
+                          "dp_size": 2,
+                          "tp_size": 8
+                  }
+            }
+        }'
+        --additional-config
+        '{"dynamic_eplb":true,"num_iterations_eplb_update":2048,"num_wait_worker_iterations":200}'
+benchmarks: