drop ascend scheduler (#4498)

Ascend scheduler was added for non chunk prefill case before, since that
the npu ops didn't work well with chunked prefill.

Now the ops with chunked prefill work better, it's time to remove the
ascend scheduler to use vLLM default scheduler.

- vLLM version: v0.11.2

---------

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>

tests/e2e/multicard/test_offline_inference_distributed.py

@@ -82,9 +82,6 @@ def test_models_distributed_DeepSeek_multistream_moe():
                     "enabled": True,
                 },
                 "enable_multistream_moe": True,
-                "ascend_scheduler_config": {
-                    "enabled": True,
-                },
                 "refresh": True,
             },
     ) as vllm_model:
@@ -154,14 +151,9 @@ def test_models_distributed_DeepSeek_W4A8DYNAMIC(model):
             quantization="ascend",
             enforce_eager=True,
             enable_expert_parallel=True,
-            additional_config={
-                "torchair_graph_config": {
-                    "enabled": False,
-                },
-                "ascend_scheduler_config": {
-                    "enabled": True,
-                }
-            },
+            additional_config={"torchair_graph_config": {
+                "enabled": False,
+            }},
     ) as vllm_model:
         vllm_model.generate_greedy(prompts, max_tokens)