WIP: Add ollama support

[HighPriest]

To run ollama models, all that we need is:

• Ollama server
• Ollama model
  • e.g. ollama pull llama3.1:8b
• Set the model name in the plugin config
  • e.g. llama3.1:8b

If no model is entered, OpenAI API is used instead.

I now recall, that openAI API also allows to choose between different APIs, so this automatic selection has to be replaced with a toggle, or some other, precise, way to distinguish if user wants to use oLLama or openAI

[HighPriest]

I am setting this PR temporarily to WIP. Got to extend the configuration with discrete distinction between running OpenAI & oLLama.