Add Retryable Test annotation

[RanVaknin]

Motivation and Context

The two modified tests are intermittently failing in our integ test code build jobs.
Since the tests in this file are using constant delay timeouts to signal an error response, and test whether they succeed / timeout, we see a race condition between the stubbed error response signaling the termination and the standard operation - both are operating on the same completable future.

A failure might look like this:

Expecting a throwable with cause being an instance of:
  software.amazon.awssdk.core.exception.ApiCallAttemptTimeoutException
but was an instance of:
  software.amazon.awssdk.services.protocolrestjson.model.ProtocolRestJsonException: Service returned HTTP status code 500 (Service: ProtocolRestJson, Status Code: 500, Request ID: null)

We could further tweak those constant timeout delay values but I don’t think it will actually solve the problem or at least not definitively. Since the existing setup is using a non-deterministic timeout delay, I believe that retrying the test when it fails with @RetryableTest(maxRetries = 3) will likely reduce failures significantly in the upstream codebuild jobs since the failure rate is already relatively low.

Testing

When I run these two locally in a @RepeatedTest suite set to 1000 repetitions, I see all of them passing with 100% success rates.

Approach to simulating the error:

For nonstreamingOperation500_notFinishedWithinTime_shouldTimeout, I tweaked the DELAY_AFTER_API_CALL_ATTEMPT_TIMEOUT from 500 to 250ms, and saw a 12% failure rate.
Adding a retryable annotation brings it down to 0.2% failure rate.

[RanVaknin]

I was not able to reproduce the test pertaining to this timeout since the value here is already low. I don't think lowering it under 50ms is going to solve anything.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud