Is your enhancement related to a problem? Please describe

There will be one default configuration out of the box. Right now, we'll be starting with llamacpp, but the goal will be to move to vLLM as soon as possible.

As a user, I should be able to configure the default inferencing runtime to be used from the settings of AI Lab.

Describe the solution you'd like

This issue will be the implementation of the UX described in #2501

Describe alternatives you've considered

No response

Additional context

No response