fix(test): correct async mock for video generation logging test

The test was failing with AuthenticationError because the mock wasn't
intercepting the actual HTTP handler calls. This caused real API calls
with no API key, resulting in 401 errors.

Root cause: The test was patching the wrong target using string path
'litellm.videos.main.base_llm_http_handler' instead of using patch.object
on the actual handler instance. Additionally, it was mocking the sync
method instead of async_video_generation_handler.

Solution: Use patch.object with side_effect pattern on the correct
async handler method, following the same pattern used in
test_video_generation_async().

Fixes test failure in PR #21277 when running with --dist=loadscope.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

tests/test_litellm/test_video_generation.py

@@ -731,50 +731,58 @@ class TestVideoLogging:
     
     @pytest.mark.asyncio
     async def test_video_generation_logging(self):
-        """Test that video generation creates proper logging payload with cost tracking."""
+        """Test that video generation creates proper logging payload with cost tracking.
+
+        Note: Uses AsyncMock with side_effect pattern for reliable parallel execution.
+        """
+        import litellm.videos.main as videos_main
+
         custom_logger = self.TestVideoLogger()
         litellm.logging_callback_manager._reset_all_callbacks()
         litellm.callbacks = [custom_logger]
-        
+
         # Mock video generation response
         mock_response = VideoObject(
             id="video_test_123",
-            object="video", 
+            object="video",
             status="queued",
             created_at=1712697600,
             model="sora-2",
             size="720x1280",
             seconds="8"
         )
-        
-        with patch('litellm.videos.main.base_llm_http_handler') as mock_handler:
-            mock_handler.video_generation_handler.return_value = mock_response
-            
+
+        # Create async mock function to return the mock_response
+        async def mock_async_handler(*args, **kwargs):
+            return mock_response
+
+        # Patch the async_video_generation_handler method on base_llm_http_handler
+        with patch.object(videos_main.base_llm_http_handler, 'async_video_generation_handler', side_effect=mock_async_handler):
             response = await litellm.avideo_generation(
                 prompt="A cat running in a garden",
                 model="sora-2",
                 seconds="8",
                 size="720x1280"
             )
-            
+
             await asyncio.sleep(1)  # Allow logging to complete
-            
+
             # Verify logging payload was created
             assert custom_logger.standard_logging_payload is not None
-            
+
             payload = custom_logger.standard_logging_payload
-            
+
             # Verify basic logging fields
             assert payload["call_type"] == "avideo_generation"
             assert payload["status"] == "success"
             assert payload["model"] == "sora-2"
             assert payload["custom_llm_provider"] == "openai"
-            
+
             # Verify response object is recognized for logging
             assert payload["response"] is not None
             assert payload["response"]["id"] == "video_test_123"
             assert payload["response"]["object"] == "video"
-            
+
             # Verify cost tracking is present (may be 0 in test environment)
             assert payload["response_cost"] is not None
             # Note: Cost calculation may not work in test environment due to mocking