[ROCm][Perf] Route the fused shared-expert gate GEMM through the platform dispatcher

vllm/model_executor/layers/fused_moe/runner/moe_runner.py

@@ -45,6 +45,7 @@
     SharedExperts,
     SharedExpertsOrder,
 )
+from vllm.model_executor.layers.utils import dispatch_unquantized_gemm
 from vllm.platforms import current_platform
 from vllm.utils.torch_utils import (
     _USE_LAYERNAME,
@@ -273,7 +274,7 @@ def __init__(
 
         # When both gates are present and FSE is enabled, fuse their
         # weight matrices into [num_experts + num_shared, hidden] so one
-        # F.linear produces combined logits. The topk kernel can then
+        # GEMM produces combined logits. The topk kernel can then
         # apply routing softmax and shared expert activation (sigmoid)
         # in a single launch.
         self._fse_fuse_gate = gate is not None and shared_expert_gate is not None
@@ -898,7 +899,9 @@ def _forward_impl(
         if self.gate is not None:
             if self._fse_fuse_gate:
                 self._maybe_fuse_gate_weights()
-                router_logits = F.linear(hidden_states, self._combined_gate_weight)
+                router_logits = dispatch_unquantized_gemm()(
+                    self, hidden_states, self._combined_gate_weight, None
+                )
             else:
                 router_logits, _ = self.gate(hidden_states)