Add Nvidia NitroGen to huggingface diffusers

**Feature request**
NVIDIA's NitroGen is a groundbreaking AI model designed to play hundreds of video games autonomously across various genres. This model, developed in collaboration with MineDojo, is trained on over 40,000 hours of publicly available gameplay videos, allowing it to perform across 1,000+ games without relying on proprietary tools or custom datasets.

**Motivation**
This model should be added to Hugging Face Transformers so that developers can use it through the HF interface and fine-tune it to play other games. Doing so will also help researchers build further on this work.

**Your contribution**
Sure, I’ll try to help with some modules, though I’m not entirely sure yet.

The issue does not name any files, tests, or entry points. Start by comparing the repository's existing model integrations and the Hugging Face Transformers interface, then determine the required NitroGen components and scope. Done means NitroGen is available through the requested Hugging Face interface and can be fine-tuned for other games.

索引モデルが issue の本文から書いたものです。