Embedding retrieval crawler (kurt)

[rngadam]

work to create embeddings:

• fetch chunk and their tokens
• submit tokens to Azure OpenAI text-embedding-ada-002 model
• set and store Referer in table link
• refactor all interactions with database (queries) in db.py

work on database:

• retrieve vector and store in pgvector in embedding

work to test embeddings:

• create api, update match_documents postgresql function
• update docs