Add Jlama inference support to rest-narration

[edeandrea]

Currently the rest-narration service is configured for either OpenAI or Azure OpenAI. Add Jlama support to this.

One caveat though is that it needs to be able to support image generation. If it doesn't, then we will have to wait on this until it does.

We'll also need to follow a similar pattern for generating separate container images per-provider. I can certainly help out if needed, but if you look in the GitHub actions you'll see how this is done.

The documentation should be updated accordingly.

[edeandrea]

@mariofusco ^^^