[bugfix][CI] Fix qwen image performance degradation w/ vllm 0.20 & CUDA 13.0 (#3352)

Signed-off-by: Huang, Zeyu <11222265+fhfuih@users.noreply.github.com>

vllm_omni/diffusion/worker/diffusion_worker.py

@@ -136,6 +136,12 @@ class DiffusionWorker:
             enable_return_routed_experts=False,
         )
         vllm_config.quant_config = self.od_config.quantization_config
+        # Since vLLM v0.20.0, IR wraps GPU ops. Set IR op priority preference to enforce GPU op fusion during wrapping.
+        # Also need to log, because vLLM internally logs another line in VllmConfig.__post_init__. Avoid confusion.
+        vllm_config.kernel_config.ir_op_priority = current_omni_platform.get_default_ir_op_priority(vllm_config)
+        logger.info(
+            "Final IR op priority after setting vLLM-Omni overrides: %s", vllm_config.kernel_config.ir_op_priority
+        )
         self.vllm_config = vllm_config
 
         # Initialize distributed environment