This PR adds a comprehensive model card for RobusTok, the image tokenizer from the paper [Image Tokenizer Needs Post-Training](https://huggingface.co/papers/2509.12474).
The updated model card includes:
- Relevant `pipeline_tag: unconditional-image-generation`, ensuring discoverability on the Hugging Face Hub.
- License information (`apache-2.0`).
- A detailed description of the model, abstract, and key highlights.
- Links to the paper, project page, and GitHub repository.
- A table showcasing model performance with direct download links.
- Comprehensive usage examples including installation, training, post-training, and inference directly sourced from the GitHub README.
- Relevant visualizations to showcase the model's capabilities.
- The official BibTeX citation.
This comprehensive update aims to provide users with all necessary information to understand, use, and reproduce the results of RobusTok.
