Create metrics and alarms for ECS tasks that stop for reasons other than scaling activities. (i.e., crashes.)
