[Core] [Bugfix] [Multimodal] Fix multimodal profiling and generation for SFT/PTQed models (#20058)

Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>

vllm/model_executor/models/fuyu.py

@@ -153,6 +153,7 @@ class FuyuMultiModalProcessor(BaseMultiModalProcessor[FuyuProcessingInfo]):
         prompt: str,
         mm_data: Mapping[str, object],
         mm_kwargs: Mapping[str, object],
+        tok_kwargs: Mapping[str, object],
     ) -> BatchFeature:
         if not mm_data:
             # Avoid warning from HF logger for text-only input
@@ -164,6 +165,7 @@ class FuyuMultiModalProcessor(BaseMultiModalProcessor[FuyuProcessingInfo]):
             prompt=prompt,
             mm_data=mm_data,
             mm_kwargs=mm_kwargs,
+            tok_kwargs=tok_kwargs,
         )
 
         image_patches = processed_outputs.get("image_patches")