ARTEMIS-5093 support configurable onMessage timeout w/closing consumer

[jbertram]

There wasn't a clear place to add documentation about this so I'm relying on JavaDoc.

[coocoo1112]

Hey sorry, I was just wondering if there was any update on this or plans to get it into a new version? Thank you so much for starting the implementation process though!

[gemmellr]

I would use a different boolean for the looping and the completion checking, it would be much clearer, and also means the set(true) doesnt then seem like dead-code (it can only ever execute here if it is already true).

The messageHandlerFinished.set(true); should also go in a try-finally so that the later assertFalse on it is guarding against unexpected-exits as well as the sole expected one being set up by the test.

You could then also wait for it to trip before the test method exits. Maybe using a CDL instead of an AtomicBoolean actually to make that succinct and efficient.

[gemmellr]

I rebased your changes, and added another temporary commit with some further changes (around the last feedback comment, and also adding a check for the tweaked timeout log message, a timeout for the overall test, reducing the close timeout a bit further).

If you are happy with the overall change we can squash them and push.