Also for optimization, we should send the MDM commands immediately during the host's MDM session/check-in, so that we don't need to issue a push notification on each command one at a time (and similarly for Windows so that we don't need to wait for the host's next MDM session).
Ensure the queue gets processed.
Some challenges have been outlined in the brainstorm document: https://github.com/fleetdm/fleet/pull/23501/files#diff-f2cc7af549e313ccbf7a6122cbcdaf299c43bf20e0eccc532c86134a3b5ba521R25-R26
Also for optimization, we should send the MDM commands immediately during the host's MDM session/check-in, so that we don't need to issue a push notification on each command one at a time (and similarly for Windows so that we don't need to wait for the host's next MDM session).