embedding with ollama

[f1-outsourcing]

I am just starting with llm and investigating a bit embedding. Currently working with continuedev on vscode. I was wondering if it would be interesting to add ollama embedding as on option? I have the impression it is advantageous as you can choose from multiple embedding models and offload to gpu.

#106