Suppress maintenance watchers during provisioning

Keep maintenance observations and automatic requests clear of active lifecycle operations and retained recovery capacity while preserving owned recovery and explicit retries.

Fixes #55

docs/maintenance.md

@@ -41,6 +41,13 @@ future runs but does not cancel requests it already created.
 All capabilities respect intentionally stopped servers. Start them explicitly
 through Provisioning before expecting maintenance to act on them.
 
+While Provisioning owns a server, Maintenance skips new health/drift observations,
+version probes, automatic updates, and scheduled restart requests for that server.
+Queued, waiting, running, and recovering operations count, including failed
+operations that retain recovery capacity. Other servers continue normally. Existing
+maintenance tracks its own operation to completion; revision verification waits
+until any later operation releases the server.
+
 ## Plutonium updates
 
 The published revision comes from the official production updater manifest.
@@ -78,7 +85,9 @@ that same revision; an operator can request an update explicitly.
 
 Daily restarts are independent of update availability. At the saved queue time,
 Maintenance requests a restart for each server whose desired state is running.
-These requests have no revision requirement. Busy servers can restart later.
+These requests have no revision requirement. Servers owned by Provisioning are
+skipped for that occurrence; the schedule advances normally. Requests already
+queued can wait for players, campaign work, or beta work to finish.
 
 Saving a schedule uses the selected timezone and follows daylight saving changes.
 Skipped times run at the first valid minute after the gap; repeated times run once