add-llm-fallbacks

Use this to make an LLM app resilient to provider failures, rate limits, timeouts, and outages. Trigger on "handle LLM API errors", "add retries/fallbacks", "the app breaks when OpenAI is down", "rate limit errors", "make my LLM calls reliable", "timeout handling". Add retries, timeouts, and model/p