[Perf][MoE] Improve MoE multistream parallel performace. (#1891)

This PR designs the shared expert multi-stream parallelism of
w8a8-dynamic-quantized MoE stage in more detail to achieve better
performance.

- vLLM version: v0.10.0
- vLLM main:
2cc571199b

Signed-off-by: whx-sjtu <2952154980@qq.com>

vllm_ascend/models/deepseek_v2.py

@@ -393,7 +393,7 @@ class CustomDeepseekV2MoE(nn.Module):
 
         # router_logits: (num_tokens, n_experts)
         router_logits = None
-        if not self.rm_router_logits:
+        if not self.rm_router_logits and not self.enable_multistream_moe:
             router_logits, _ = self.gate(hidden_states)
 
         experts_hidden_states = self.experts(