When an isolator kills a task, the reason is unknown. As part of MESOS-1830, the reason is set to a general one but ideally we would have the termination reason to pass through to the status update.
mesos should send docker failure messages to scheduler
bogus docker images result in bad error message to scheduler
Remove killed from Termination protobuf message.
Expose TASK_FAILED reason to Frameworks.
Expose master stats differentiating between master-generated and slave-generated LOST tasks
Support multiple reasons in status update message.
Use mesos' disk quota enforcement
Propagate Isolator::prepare() failures to the framework
Status update for non-terminal tasks when a container terminates