Description of MB-35748 does not explain the problem

Currently the description of this MB says something like "Handle trap_exit better", this does not really tell the user what the problem is or what the impacts could be, which is what should be in the release notes.

I suggest having something similar to the following:

Resolved an issue where an orchestration process would get stuck shutting down within the Cluster Manager.
This caused topology changes within the cluster (such as rebalances or failovers) to consistently fail with an error around having `no_quorum` until the Cluster Manager on this node was restarted.