Add Cosmos3 action modality (#4102)

Signed-off-by: Maciej Bala <mbala@nvidia.com>
Signed-off-by: Bartosz Stefaniak <bstefaniak@nvidia.com>
Signed-off-by: lishunyang12 <lishunyang12@163.com>
Co-authored-by: Maciej Bala <mbala@nvidia.com>
Co-authored-by: lishunyang12 <lishunyang12@163.com>
Co-authored-by: SYLAR <125541396+lishunyang12@users.noreply.github.com>

vllm_omni/entrypoints/openai/api_server.py

@@ -2545,7 +2545,7 @@ async def _run_video_generation_job(
     started_at = time.perf_counter()
     output_path = None
     try:
-        video_bytes, stage_durations, peak_memory_mb = await handler.generate_video_bytes(
+        video_bytes, stage_durations, peak_memory_mb, action = await handler.generate_video_bytes(
             request, video_id, reference_image=reference_image
         )
 
@@ -2563,6 +2563,7 @@ async def _run_video_generation_job(
                 "inference_time_s": time.perf_counter() - started_at,
                 "stage_durations": stage_durations,
                 "peak_memory_mb": peak_memory_mb,
+                "action": action,
             },
         )
     except (EngineGenerateError, EngineDeadError) as exc:
@@ -2774,7 +2775,7 @@ async def create_video_sync(
     raw_request.state.request_metadata = RequestResponseMetadata(request_id=request_id)
     started_at = time.perf_counter()
     try:
-        video_bytes, stage_durations, peak_memory_mb = await asyncio.wait_for(
+        video_bytes, stage_durations, peak_memory_mb, _action = await asyncio.wait_for(
             handler.generate_video_bytes(request, request_id, reference_image=reference_image),
             timeout=VIDEO_SYNC_TIMEOUT_S,
         )