OpenAI Batch API Support

[thomaswitt]

I think the evolution of ruby_llm in the last months is great, thanks for all the effort.

The only thing which stops me on switching on several projects is the lack of supporting the batch processing API (https://platform.openai.com/docs/guides/batch) which makes processing large non-timecritical datasets for embeddings or analysis 50% cheaper. I use it a lot of times in different projects.

Any plans on the support?