👓 feat: Vision Support for Assistants

[danny-avila]

Summary

Introduces vision support for Assistants.

This may end up being natively supported by OpenAI but wanted to introduce at least an alternative method, as will be done for retrieval.

Their docs state they intend to do this at some point:

Messages can contain text, images, or files. At the moment, user-created Messages cannot contain image files but we plan to add support for this in the future. source

In anticipation of this, I've name the tool image_vision as opposed to just vision

Todo's (in another PR)

• Tooltips over Assistant features to better explain them
• Azure Support (may or may not work at the moment depending on config, needs robust approach)
• Allow uploading images both to Host and OpenAI (for Code Interpreter)
  • Still deliberating how this might work

Other changes

• Fixed a React key error that only occurred when the server was shut down before an assistant run could be initialized (edge case)
• Refactors some of the handling of necessary API calls per-initialization of Assistant run in parallel now

Change Type

• New feature (non-breaking change which adds functionality)

Checklist

• My code adheres to this project's style guidelines
• I have performed a self-review of my own code
• I have commented in any complex areas of my code
• I have made pertinent documentation changes
• My changes do not introduce new warnings
• I have written tests demonstrating that my changes are effective or that my feature works
• Local unit tests pass with my changes
• Any changes dependent on mine have been merged and published in downstream modules.
• New documents have been locally validated with mkdocs