[refactor]support gatingtopk operator generalization (#2958)

### What this PR does / why we need it?

Past：
npu_moe_gating_top_k can only support 'group_count=256' pattern

Now：
1、npu_moe_gating_top_k support all size of group_count
2、the functionality of `torch_npu.npu_moe_gating_top_k_softmax` are
included in `torch_npu.npu_moe_gating_top_k`

CANN: depends on 8.3.RC1

Performance：
1. GLM4.5-w8a8, TPS improve 6%
2. Qwen3, the same as before

- vLLM version: v0.11.0
- vLLM main:
2918c1b49c

Signed-off-by: 1092626063 <1092626063@qq.com>

vllm_ascend/ascend_forward_context.py

@@ -98,6 +98,7 @@ def set_ascend_forward_context(
         ep_size = (get_ep_group().world_size if
                    vllm_config.parallel_config.enable_expert_parallel else 1)
 
+        # fused_moe_state is used in torchair, it will be deleted along with torchair
         is_deepseek_v3_r1 = hasattr(
             vllm_config.model_config.hf_config, 'n_routed_experts'
         ) and vllm_config.model_config.hf_config.n_routed_experts == 256