When a cluster is deleted we should attempt to clean up the staging bucket associated with that cluster. Instead of immediately deleting it, we should set the lifecycle on that bucket so that it's deleted 1 week after the cluster is deleted. There is often useful information in the staging bucket that can help us debug why clusters failed to start so we don't want to lose that.
We'll also need to figure out how to clean up any orphaned staging buckets that may be lingering.