Support Pangu Pro MoE model (#1204)

### What this PR does / why we need it?
Support Pangu Pro MoE model (https://arxiv.org/abs/2505.21411)

### Does this PR introduce _any_ user-facing change?
Yes, new model supported

### How was this patch tested?
Test locally

---------

Signed-off-by: angazenn <zengyanjia@huawei.com>
Co-authored-by: angazenn <zengyanjia@huawei.com>

vllm_ascend/ops/fused_moe.py

@@ -759,6 +759,7 @@ def select_experts(
     custom_routing_function: Optional[Callable] = None,
     scoring_func: str = "softmax",
     e_score_correction_bias: Optional[torch.Tensor] = None,
+    global_num_experts: Optional[torch.Tensor] = None
 ) -> tuple[torch.Tensor, torch.Tensor]:
     """
     Select top-k experts based on router logits.
@@ -826,7 +827,8 @@ def select_experts(
             hidden_states=hidden_states,
             gating_output=router_logits,
             topk=top_k,
-            renormalize=renormalize)
+            renormalize=renormalize,
+            global_num_experts=global_num_experts)
         # Required by npu_moe_init_routing
         topk_ids = topk_ids.to(torch.int32)
         return topk_weights, topk_ids