brig check should be ported from 1.x to 2.x

In v1.x, this command (like everything else) would have required direct access to the cluster.

In 2.x, the right thing to do is add an endpoint to the system API that checks on all components and returns their statuses. This was v2.x users who may _not_ have direct access to the underlying cluster can still make determinations about the health of the brigade instance.

Start by locating the v1.x `brig check` command and the v2.x system API entry point. Trace how component health is currently determined, then define the endpoint response and connect the command to it. Done means users without direct cluster access can run `brig check` and receive the status of all components.

Written by the indexing model from the issue text.