SERVER-110796 and associated tickets introduced RangeDeletionTaskTracker, which blocks the range deleter until all tasks are recovered (see SERVER-110796 for why this was necessary).

To some extent, this was motivated by the ad hoc nature of migration coordinator recovery prior to the introduction of the MoveRangeCoordinator in SPM-4061. Now that migrations are PrimaryOnlyServices, we may be able to simplify this and have the range deleter directly wait for MoveRangeCoordinators to complete, eliminating the need for the RangeDeletionTaskTracker entirely.

Even if the RangeDeletionTaskTracker is still necessary, we should consider if there's a better place to invoke the job recovery notifier than ShardingCoordinatorService (see SERVER-130215).