Implement async mutation to store AI CI config Chats

What does this MR do and why?

Adds a service that persists messages from user and the ai. Hooks that service into an existing mutation.

Each time we have to pass the full conversation history back to the AI for it to be contextually aware.

WE need to delete old messages as well to stay under token model limits.

We do this operation async since the request can take a long time.

Issue: https://gitlab.com/gitlab-org/gitlab/-/issues/407739

How to set up and validate locally

!117924 (closed)

MR acceptance checklist

This checklist encourages us to confirm any changes have been analyzed to reduce risks in quality, performance, reliability, security, and maintainability.

• I have evaluated the MR acceptance checklist for this MR.