[ISSUE #1588] Fix 1588 retry doesn't work when sending msg in async mode and the broker is down

[Aaron-He]

Fix bug: retry doesn't work when sending msg in async mode and the
broker is down.

What is the purpose of the change

Fix bug: retry doesn't work when sending msg in async mode and the
broker is down.

Brief changelog

When Sending msg in async mode to a down broker, RemotingConnectException will be thrown since the channel is null. Catch the exception and call onExceptionImpl method to retry sending msg.

Verifying this change

YES.

Follow this checklist to help us incorporate your contribution quickly and easily. Notice, it would be helpful if you could finish the following 5 checklist(the last one is not necessary)before request the community to review your PR.

• Make sure there is a Github issue filed for the change (usually before you start working on it). Trivial changes like typos do not require a Github issue. Your pull request should address just this issue, without pulling in other changes - one PR resolves one issue.
• Format the pull request title like [ISSUE #123] Fix UnknownException when host config not exist. Each commit in the pull request should have a meaningful subject line and body.
• Write a pull request description that is detailed enough to understand what the pull request does, how, and why.
• Write necessary unit-test(over 80% coverage) to verify your logic correction, more mock a little better when cross module dependency exist. If the new feature or significant change is committed, please remember to add integration-test in test module.
• Run mvn -B clean apache-rat:check findbugs:findbugs checkstyle:checkstyle to make sure basic checks pass. Run mvn clean install -DskipITs to make sure unit-test pass. Run mvn clean test-compile failsafe:integration-test to make sure integration-test pass.
• If this contribution is large, please file an Apache Individual Contributor License Agreement.

[coveralls]

Coverage increased (+0.02%) to 50.664% when pulling 57ef806 on Aaron-He:develop into f8f6fbe on apache:develop.

[xiangwangcheng]

This solution can not cover the situation when sending message timeout before the sendMessageAsync is invoked.
As I mentioned in the issue, reuse the loop retry logic would be simpler and clearer.

This make us have two seperated retry logic, one is in charge of resending messages when failed in the sending phrase, the other is in charge of resending when failed in the response-process phrase.

[Aaron-He]

@xiangwangcheng Good solution.