Add FalkordbRM Retriever Class for Enhanced Querying and Embedding Retrieval

[kingtroga]

This pull request introduces the FalkordbRM retriever class, designed to integrate FalkorDB for efficient querying and passage retrieval based on embeddings. The class supports both local and cloud-based FalkorDB sessions and uses OpenAI's embedding services by default

Key Changes:

1. FalkordbRM Retriever Class
2. Embedding Provider Integration
3. Error Handling
4. Customization Options

[gkorland]

@okhat @arnavsinghvi11 can you please merge?

[arnavsinghvi11]

Thanks @kingtroga @gkorland for this PR! Seems like some merge conflicts should be resolved if you pull from current main! should pass the failing test as well

also added a comment to support pip install dspy-ai[falkordb] correctly.

[kingtroga]

@arnavsinghvi11 I've made the necessary changes. Please try the merge again 🙏🏾

[kingtroga]

@arnavsinghvi11

[arnavsinghvi11]

Thanks @kingtroga !

we're currently planning an internal refactor of how we support RMs in DSPy and hence will leave this PR open to avoid any future code reversions. However, if the PR requires merging to be usable in DSPy, we'll do so! Please let us know

[kingtroga]

Hi @arnavsinghvi11 , thanks for the update! This PR does need merging to be usable in DSPy. Let me know if there's anything I can do to help with it.

[arnavsinghvi11]

Thanks @kingtroga !