Add support for someone to be able to use a Locally hosted LLM.

• The LocalLLM must have an API wrapper like GPT4ALL
• This will not allow an arbitrary LLM running inferences to just interact with the system, we will not be building API wrappers for LLMs.

References for easier integration:
https://docs.gpt4all.io/
LocalAI
Replicant
OLLama
LLMStudio
HuggingFace