This PR improves the model card by:
- Adding the `image-text-to-text` pipeline tag.
- Specifying `transformers` as the library name (compatible via `Qwen2_5_VLForConditionalGeneration`).
- Linking the model to its research paper: [Beyond Clicking: A Step Towards Generalist GUI Grounding via Text Dragging](https://huggingface.co/papers/2601.06031).
- Adding the `base_model` link to `xlangai/Jedi-3B-1080p`.
- Adding tags for `gui-grounding` and `text-dragging`.
- Organizing links to the project page and GitHub repository.