[MUSA] Fix SwiGLU to use fused op (#6364)

Signed-off-by: Xiaodong Ye <xiaodong.ye@mthreads.com>

vllm_omni/diffusion/layers/activation.py

@@ -22,7 +22,7 @@ class SiluAndMul(CustomOp):
         return F.silu(x[..., :d]) * x[..., d:]
 
     def forward_cuda(self, x: torch.Tensor) -> torch.Tensor:
-        if x.device.type != "cuda":
+        if x.device.type == "cpu":
             return self.forward_native(x)
         d = x.shape[-1] // 2
         output_shape = x.shape[:-1] + (d,)