Fix gap in sentry monitoring
We've had a couple of instances where sentry stops processing events due to kafka lag / offset errors.
Our dashboards do show the events processed dropping to 0, but we do not receive any alerts, or have any alert configured to notify us here.
This leads to a gap in our visibility/monitoring of sentry for these type of events, and can go un-noticed for long periods of times until someone reports this to us
We need to remediate this.
Additionally we need to investigate why we can't auto recover here, are we not using auto offset reset? These events often follow a larger incident where Gitlab throws many more exceptions. This leads to the lag in kafka building due to sentry falling behind / being unable to process the messages fast enough.