Implement LlamaGen for Image Generation

### Feature request Add support for LlamaGen, an autoregressive image generation model, to the Transformers library. LlamaGen applies the next-token prediction paradigm of large language models to visual generation. Paper: https://arxiv.org/abs/2406.06525 Code: https://github.com/FoundationVision/LlamaGen Key components to implement: 1. Image tokenizer 2. Autoregressive image generation model (based on Llama architecture) 3. Class-conditional and text-conditional image generation 4. Classifier-free guidance for sampling ### Motivation LlamaGen demonstrates that vanilla autoregressive models without vision-specific inductive biases can achieve state-of-the-art image generation performance. Implementing it in Transformers would enable easier experimentation and integration with existing language models. ### Your contribution I can help by contributing to this model, and provide examples and detailed explanations of the model architecture and training process if needed.