[Kernels] Overlap shared experts with send/recv (#23273)

Signed-off-by: Bill Nell <bnell@redhat.com>

vllm/model_executor/layers/quantization/rtn.py

@@ -3,7 +3,7 @@
 # Copyright © 2025, Oracle and/or its affiliates.
 
 import os
-from typing import Any, Callable, Optional
+from typing import Any, Callable, Optional, Union
 
 import torch
 import torch.nn.functional as F
@@ -291,7 +291,7 @@ class RTNMoEMethod(FusedMoEMethodBase):
         expert_load_view: Optional[torch.Tensor] = None,
         logical_to_physical_map: Optional[torch.Tensor] = None,
         logical_replica_count: Optional[torch.Tensor] = None,
-    ) -> torch.Tensor:
+    ) -> Union[torch.Tensor, tuple[torch.Tensor, torch.Tensor]]:
         assert self.fused_experts is None
 
         if enable_eplb: