Implement Support for HyDE

[VivaLaPanda]

Using the vector store DBs with the default OpenAI embeddings as pretty meh results, because the OpenAi embeddings aren't really optimized for search. We could use HyDE to improve this, and langchain already has this implemented: https://langchain.readthedocs.io/en/latest/modules/utils/combine_docs_examples/hyde.html

I've spent some time hacking around, but I'm not sure where the most natural place to inject this into the gpt_index flow is.