Revert "drop ascend scheduler" (#4580)

Reverts vllm-project/vllm-ascend#4498
- vLLM version: v0.11.2
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.2

tests/e2e/nightly/features/test_mtpx_deepseek_r1_0528_w8a8.py

@@ -73,7 +73,11 @@ async def test_models(model: str, mode: str) -> None:
         "VLLM_RPC_TIMEOUT": "3600000",
         "VLLM_EXECUTE_MODEL_TIMEOUT_SECONDS": "3600000"
     }
-    additional_config: dict[str, Any] = {}
+    additional_config: dict[str, Any] = {
+        "ascend_scheduler_config": {
+            "enabled": False
+        },
+    }
     speculative_config = {
         "num_speculative_tokens": 2,
         "method": "deepseek_mtp"

tests/e2e/nightly/features/test_prefix_cache_deepseek_r1_0528_w8a8.py

@@ -74,6 +74,9 @@ async def test_models(model: str) -> None:
         "PYTORCH_NPU_ALLOC_CONF": "expandable_segments:True",
     }
     additional_config = {
+        "ascend_scheduler_config": {
+            "enabled": False
+        },
         "torchair_graph_config": {
             "enabled": True,
             "enable_multistream_moe": False,

tests/e2e/nightly/features/test_prefix_cache_qwen3_32b_int8.py

@@ -68,7 +68,12 @@ aisbench_cases75 = [{
 async def test_models(model: str) -> None:
     port = get_open_port()
     env_dict = {"TASK_QUEUE_ENABLE": "1", "HCCL_OP_EXPANSION_MODE": "AIV"}
-    additional_config = {"enable_weight_nz_layout": True}
+    additional_config = {
+        "ascend_scheduler_config": {
+            "enabled": False
+        },
+        "enable_weight_nz_layout": True
+    }
     server_args = [
         "--quantization", "ascend", "--reasoning-parser", "qwen3",
         "--tensor-parallel-size", "4", "--port",

tests/e2e/nightly/features/test_qwen3_32b_int8_a3_feature_stack3.py

@@ -83,7 +83,8 @@ async def test_models(model: str, tp_size: int) -> None:
         "0.9", "--block-size", "128", "--max-num-seqs", "256",
         "--enforce-eager", "--max-model-len", "35840",
         "--max-num-batched-tokens", "35840", "--additional-config",
-        '{"enable_weight_nz_layout":true}', "--compilation-config",
+        '{"ascend_scheduler_config":{"enabled":true},"enable_weight_nz_layout":true}',
+        "--compilation-config",
         '{"cudagraph_mode":"FULL_DECODE_ONLY", "cudagraph_capture_sizes":[1,8,24,48,60]}'
     ]
     with RemoteOpenAIServer(model,