fix: Add 'image-text-to-text' to TASK_MAPPING

Author: saattrupdan

Diff for: src/transformers/modelcard.py

@@ -33,6 +33,7 @@
     MODEL_FOR_CTC_MAPPING_NAMES,
     MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING_NAMES,
     MODEL_FOR_IMAGE_SEGMENTATION_MAPPING_NAMES,
+    MODEL_FOR_IMAGE_TEXT_TO_TEXT_MAPPING_NAMES,
     MODEL_FOR_MASKED_LM_MAPPING_NAMES,
     MODEL_FOR_OBJECT_DETECTION_MAPPING_NAMES,
     MODEL_FOR_QUESTION_ANSWERING_MAPPING_NAMES,
@@ -70,6 +71,7 @@
     "audio-classification": MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING_NAMES,
     "automatic-speech-recognition": {**MODEL_FOR_CTC_MAPPING_NAMES, **MODEL_FOR_SPEECH_SEQ_2_SEQ_MAPPING_NAMES},
     "zero-shot-image-classification": MODEL_FOR_ZERO_SHOT_IMAGE_CLASSIFICATION_MAPPING_NAMES,
+    "image-text-to-text": MODEL_FOR_IMAGE_TEXT_TO_TEXT_MAPPING_NAMES,
 }
 
 logger = logging.get_logger(__name__)