Implement pluggable vector indexing / seach engines

[crossjam]

I'd like to experiment with multiple different vector search libraries such as zvec, chromadb, and usearch.

It should be possible to use the Python pluggy library to implement pluggable framework to integrate particular engines. This should follow the model in Simon Willison's llm library

Here's what his documentation says about plugins

LLM plugins can enhance LLM by making alternative Large Language Models available, either via API or by running the models locally on your machine.

Plugins can also add new commands to the llm CLI tool.

The overview of plugin hooks provides an example design.

Here's the source code for the llm plugin module