[bugfix] fix mtp accept rate (#5093)

### What this PR does / why we need it?
1. now, npu_model_runner reuses gpu_model_runner, this pr deletes some
attrs already defined in gpu_model_runner
2. fix mtp accept rate by disabling in_profile_run
3. remove redundant moe method selection logic
4. Reverts vllm-project/vllm-ascend#5082, which broke CI in
https://github.com/vllm-project/vllm-ascend/actions/runs/20266314048/job/58190426832?pr=5088

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

### How was this patch tested?
vLLM version: v0.12.0
vLLM main:
ad32e3e19c

vLLM version: v0.12.0
vLLM main:
ad32e3e19c

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: zhenwenqi2024 <zhenwenqi_2022@qq.com>
Signed-off-by: Mengqing Cao <cmq0113@163.com>
Co-authored-by: Mengqing Cao <cmq0113@163.com>

vllm_ascend/ascend_forward_context.py

@@ -64,7 +64,7 @@ def set_ascend_forward_context(
             get_moe_comm_method
         moe_comm_type = select_moe_comm_method(num_tokens, vllm_config)
         # TODO: remove this after moe_comm_type selection logic is finalized
-        if in_profile_run and is_mtp_model:
+        if is_mtp_model:
             moe_comm_type = (MoECommType.ALLTOALL if moe_comm_type
                              == MoECommType.FUSED_ALLTOALL else moe_comm_type)
         forward_context.moe_comm_type = moe_comm_type
@@ -298,8 +298,6 @@ def select_moe_comm_method(num_tokens: int,
                          if fused_all2all_enable else MoECommType.ALLTOALL)
     else:
         raise ValueError(f"Unsupported soc_version: {soc_version}")
-    moe_comm_type = (MoECommType.ALLTOALL if moe_comm_type
-                     == MoECommType.FUSED_ALLTOALL else moe_comm_type)
     # PanguProMoE only supports allgather
     if model_type == "PanguProMoE":
         moe_comm_type = MoECommType.ALLGATHER