Refactor src and chat service

[MichaelClifford]

In an attempt to make this repo more modular while retaining the use of some shared resources, I created a chatbot directory that contains an improved version of what was originally the only application in this repo, the gradio chat bot.

• This new directory mimics the template used by summarizer with a model_service and ai_applications subdirectories where we store the code for serving the local llm applications as well as demo interaction apps.
• Rewrote the repo README.md to reflect the fact that it now contains multiple (and growing) local LLM examples.
• Created a llamacpp_utils.py file to contain the shared functions across the different examples.
• Removed a number of files that were no longer necessary.