feat: TRL Integration

[bxyu-nvidia]

Use cases, pain points, and background Why should we do this? Why is this needed or wanted?

Description: What should we do?

Design: What files should be touched? What logic should be written?

Out of scope: What are some items that this issue could be mistaken to cover that this issue should explicitly NOT cover?

Acceptance Criteria:

• [ ] Individual items that need to be finished in order for this issue to be considered completed

[cmunley1]

TRL has a custom rollout function and vllm server mode that makes the integration easier. The vllm server is not a typical AsyncLLMEngine, it does not have openai chat completions/responses endpoints currently, we may have to add support for that https://github.com/huggingface/trl/issues/4602.