moe_gating_top_k (#5271)

1. What this PR does / why we need it?
This PR supports the moe_gating_top_k operator, which enables
post-positioned renormalization (renorm) on the basis of softmax.
2. Does this PR introduce any user-facing change?
No user-facing changes are required.
3. How was this patch tested?
This patch was tested with the test_npu_moe_gating_top_k test case.
vLLM version: release/v0.13.0
vLLM main:
ad32e3e19c

---------

Signed-off-by: ZCG12345 <2097562023@qq.com>
Signed-off-by: zzzzwwjj <34335947+zzzzwwjj@users.noreply.github.com>
Co-authored-by: zzzzwwjj <34335947+zzzzwwjj@users.noreply.github.com>

tests/ut/model_loader/netloader/test_netloader_elastic.py

@@ -311,7 +311,7 @@ def test_client_handler_mismatch(server_config):
         mismatch_data = {
             "label": "JOIN",
             "content": {
-                "device_id": 1,  # 不匹配的ID
+                "device_id": 1,  # Mismatched ID
                 "model_path": "/wrong/model",
                 "tp": 2,
                 "pp": 2,

tests/ut/worker/test_worker_v1.py

@@ -670,7 +670,7 @@ class TestNPUWorker(TestBase):
             (5000, 10000),
         ]
 
-        # 创建 worker mock
+        # Create worker mock
         with patch.object(NPUWorker, "__init__", lambda x, **kwargs: None):
             worker = NPUWorker()
             worker.init_npu_memory = 8500