Add t5gemma2 to image-text-to-text category.

Author: bzhangGo

src/transformers/models/auto/modeling_auto.py

@@ -1062,6 +1062,7 @@ class _BaseModelWithGenerate(PreTrainedModel, GenerationMixin):
         ("qwen3_vl_moe", "Qwen3VLMoeForConditionalGeneration"),
         ("shieldgemma2", "Gemma3ForConditionalGeneration"),
         ("smolvlm", "SmolVLMForConditionalGeneration"),
+        ("t5gemma2", "T5Gemma2ForConditionalGeneration"),
         ("udop", "UdopForConditionalGeneration"),
         ("video_llama_3", "VideoLlama3ForConditionalGeneration"),
         ("vipllava", "VipLlavaForConditionalGeneration"),