Don't let potential failures in message consumers cause dev mode shutdown to be blocked permanently

[jaikiran]

The commit here makes sure that the unregisterMessageConsumers counts down the latch irrespective of whether or not the message consumer failed. Additionally, a failure of the message consumer to unregister is now logged (at WARN) during the shutdown.

While I was at this, I noticed that the registration of these message consumers too had a potential to block forever awaiting a latch countdown, if the message consumer fails for whatever reason. So I've updated that part of the code too, to count down the latch irrespective of whether or not the registration failed. A failure to register is then thrown back as an exception to prevent the deployment from succeeding.

[quarkus-bot]

Test Failures

⚙️ jvm-linux-jdk11

📦 extensions/mongodb-client/deployment

# Tests:    11
+ Success:  10
- Failures: 0
- Errors:   1
! Skipped:  0

❌ io.quarkus.mongodb.LegacyNamedMongoClientConfigTest - see on GitHub

java.util.concurrent.RejectedExecutionException
	at org.jboss.threads.RejectingExecutor.execute(RejectingExecutor.java:38)
	at org.jboss.threads.EnhancedQueueExecutor.rejectShutdown(EnhancedQueueExecutor.java:2045)
	at org.jboss.threads.EnhancedQueueExecutor.execute(EnhancedQueueExecutor.java:767)
	at io.smallrye.mutiny.infrastructure.MutinyScheduler$DecoratedRunnableTask.run(MutinyScheduler.java:58)
	at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128)
	at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628)
	at java.base/java.lang.Thread.run(Thread.java:834)
	Suppressed: java.util.concurrent.RejectedExecutionException: Executor is being shut down
		at org.jboss.threads.EnhancedQueueExecutor.rejectShutdown(EnhancedQueueExecutor.java:2047)
		... 5 more

[mkouba]

Looks good.

[jaikiran]

I've updated the commit message and the PR description to remove references to #16230, since although this PR does fix a potential issue, it may not fix what's described in that issue, for reasons noted in #16230 (comment)