Fix SshClient thread leak in EC2SSHLauncher.launchRemotingAgent()

[olamy]

• each launchRemotingAgent() creates an Apache MINA SshClient (which spawns internal thread pools nio acceptor/selectors) but only registers cleanup via a Channel.Listener callback at the end when happens of the method. If any exception occurs before computer.setChannel() is reached, the client is never cleaned up and so leaking many threads.
• With Windows SSH agents that retry frequently during boot, this produces thousands of leaked sshd-SshClient threads.

Signed-off-by: Olivier Lamy olamy@apache.org

Testing done

Submitter checklist

• Make sure you are opening from a topic/feature/bugfix branch (right side) and not your main branch!
• Ensure that the pull request title represents the desired changelog entry
• Please describe what you did
• Link to relevant issues in GitHub or Jira
• Link to relevant pull requests, esp. upstream and downstream changes
• Ensure you have provided tests that demonstrate the feature works or the issue is fixed

[olamy]

this is conservative option for #2008

[jglick]

Looks reasonable. Has this been tested in a real controller connecting to real AWS, reproducing the stated error and confirming that this addresses it?