Improve AI Bot feedback

[aj-codecov]

Is your feature request related to a problem? Please describe.

• “Engineers don’t know if the reviewer is working/not working/thinks everything is fine”
• “Engineers think that it’s not working because @tting the bot doesn’t work like normal @tting”
• “AI PR Review is flaky”

Describe the solution you'd like
When a user @-s the bot we need to confirm we have received the command, let them know we're working it, and give a final outcome of a review, an error, or an "LGTM" type message

Describe alternatives you've considered
This is table stakes for this sort of feature, doing nothing isn't really an option.

Additional context
The Seer migration will give us some of this for free but the task here is to identify what that looks like when it arrives and then fine tune for the additional scenarios once we understand what else is needed