Instance health indicator in usage ping

Overview

Instance usage ping provides GitLab with useful information to help us understand how folks are using the product. We don't, however, send any metrics that give us a sense of the "health" of an instance. How much use is it getting? Is the instance being actively used, or is it slowly dying?

Adding indications of an instance's "health" might be useful for GitLab, as it might enable us to understand usage patterns that differentiate a super healthy and active instance vs. one less so.

Proposal

Consider engagement ratio as a "health metric", defined by Weekly Active Users / Monthly Active Users. Add these to the usage ping.