This PR adds a comprehensive model card for `CodePlot-CoT`. It includes:
- Linking to the paper [CodePlot-CoT: Mathematical Visual Reasoning by Thinking with Code-Driven Images](https://huggingface.co/papers/2510.11718).
- `library_name: transformers` (evidence from `config.json`) and `pipeline_tag: image-to-text` for discoverability.
- `license: mit` (explicitly stated in GitHub README).
- Links to the project page and GitHub repository.
- A detailed introduction, model overview, and released data/models.
- An explicit sample usage section with installation and inference code snippets directly from the GitHub README.
- Relevant tags, datasets, and languages for improved searchability.
This enhances the discoverability and usability of the model significantly.
