`ceph-salt update` reports failure when installing updates to salt itself

When you run `ceph-salt update`, it goes off and installs all available package updates. This may include updates to salt itself. If there are salt updates, the salt minions will all usually be restarted as part of the update process. At this point, the active `ceph-salt update` will return failure, although there are still package updates being run in the background by some salt minions that haven't restarted yet. While this is going on, you'll see "An instance is already running. Exiting the Salt Minion" in the minion logs, because the minions are trying to restart while busy. Eventually the updates will finish and the minions will restart and everything will be fine again. You should probably run `ceph-salt update` again for good measure once everything has settled down.

It would be ideal if `ceph-salt update` could somehow detect that salt updates are available and handle this case more helpfully. One option may be to do what DeepSea does, which is to tell the user that there are salt updates, and request the user run a specific command to install only those updates first before installing everything else.

The payload names only the `ceph-salt update` command; first locate its Python entry point and update orchestration, then reproduce an update that includes Salt itself while watching minion logs. Done means the command no longer reports a misleading failure and gives the operator a clear way to handle Salt updates before remaining packages; verify behavior after minions restart.

Written by the indexing model from the issue text.