Add pipeline tag and library name to model card

This PR enhances the model card for VisCoder2-7B by adding key metadata:

- `pipeline_tag: image-text-to-text`: This helps users discover the model under multimodal generation tasks on the Hugging Face Hub, as it generates visualization code that produces images from text instructions.
- `library_name: transformers`: The `config.json` indicates compatibility with `transformers` (e.g., `"model_type": "qwen2"`, `"transformers_version": "4.51.3"`), enabling the automated inference widget and "How to use" code snippets on the Hub.

Please review these additions.

README.md

@@ -1,13 +1,15 @@
 ---
-license: apache-2.0
+base_model:
+- Qwen/Qwen2.5-Coder-7B-Instruct
 datasets:
 - TIGER-Lab/VisCode-Multi-679K
 language:
 - en
-base_model:
-- Qwen/Qwen2.5-Coder-7B-Instruct
+license: apache-2.0
 tags:
 - code
+pipeline_tag: image-text-to-text
+library_name: transformers
 ---
 
 # VisCoder2-7B