Terminate ns_vbucket_mover synchronously when rebalance is stopped.

This code isn't executed when a rebalance is stopped since we run rebalance_body via an async and the async code converts all terminate requests Reason to a "shutdown".

In this particular case an user intitated shutdown Reason "{shutdown, stop}" is converted to a "shutdown".

Here are logs from a rebalance stopped in the baseline code and this log isn't printed.

fd "debug.log" | xargs -I {} sh -c "rg \"Got rebalance stop\" {}"

Rebalance stopped via user log in ns_orchestrator:

fd "debug.log" | xargs -I {} sh -c "rg \"Rebalance stopped\" {}"

29150:[user:info,2022-10-07T11:31:04.021-07:00,n_0@192.168.86.150:<0.4997.0>:ns_orchestrator:log_rebalance_completion:1448]Rebalance stopped by user.

Based on a discussion with Artem - terminating ns_vbucket_mover async is dangerous and can potentially lead to data-loss, moving the fix to match on shutdown instead of {shutdown, stop}.

Artem Stemkovski  1:03 PMThe consequences of the orphaned mover doing something could be quite serious. Like data corruption