Improve model card: Add metadata and GitHub link

This PR updates the model card for the COOPER model to enhance its discoverability and provide more complete information:

- Adds `pipeline_tag: image-to-image` to the YAML metadata, reflecting the model's core capabilities in perception enhancement (depth estimation, semantic segmentation).
- Adds `library_name: transformers` to the YAML metadata, as evidenced by the `pip install -e ./transformers-4.54.0` instruction in the "Quick Start" guide. This enables the automated "Use in Transformers" code snippet on the Hub.
- Adds `license: apache-2.0` to the YAML metadata to specify the model's license.
- Updates the existing paper link in the markdown content to point to the Hugging Face paper page: https://huggingface.co/papers/2512.04563.
- Adds an explicit link to the GitHub repository (`https://github.com/zhangzef/COOPER`) in the header for easier access to the code.

These changes will make the model card more informative and user-friendly.

Please review and merge if these improvements align with the repository's goals.

README.md

@@ -1,7 +1,14 @@
+---
+pipeline_tag: image-to-image
+library_name: transformers
+license: apache-2.0
+---
+
 # COOPER 🧭
 
 <p align="center">
-  📄 <a href="https://arxiv.org/pdf/2512.04563">Paper</a> |
+  📄 <a href="https://huggingface.co/papers/2512.04563">Paper</a> |
+  💻 <a href="https://github.com/zhangzef/COOPER">Code</a> |
   🤖 <a href="https://huggingface.co/Starrrrrry/COOPER">COOPER Model</a> |
   🧠 <a href="https://huggingface.co/Starrrrrry/COOPER-AMG">COOPER-AMG Model</a> |
   📂 <a href="https://huggingface.co/datasets/Starrrrrry/COOPER_Train_Set">COOPER Training Data</a>