Opinionated request throttling utility to help manage token and request rate limits to the OpenAI API.
