Update Llama3VILAInferenceOperator to support optional image_metadata parameter type

- Changed the type hint for the image_metadata parameter from Dict to Dict | None, allowing for clearer handling of optional values in the inference operator.

Signed-off-by: Victor Chang <vicchang@nvidia.com>

Author: mocsharp

monai/deploy/operators/llama3_vila_inference_operator.py

@@ -186,7 +186,7 @@ def _create_json_result(
         text_response: str,
         request_id: str,
         prompt: str = None,
-        image_metadata: Dict = None,
+        image_metadata: Dict | None = None,
     ) -> Dict[str, Any]:
         """Create a JSON result from the text response."""
         result = {