GEODE-10556: Fix race condition in PulseSecurityWithSSLTest

[JinwooHwang]

Problem

PulseSecurityWithSSLTest intermittently failed on GitHub Actions with authentication errors (BAD_CREDS) despite using correct credentials. Tests consistently passed locally but failed in CI where JMX/Management Service initialization is slower.

Root Cause: Race condition where authentication attempts occurred before ManagementService and MemberMXBean were fully initialized.

Solution

Added synchronization using GeodeAwaitility.await() to ensure ManagementService is ready before authentication:

ManagementService service = ManagementService.getExistingManagementService(locator.getLocator().getCache());
await().untilAsserted(() -> assertThat(service.getMemberMXBean()).isNotNull());

Applied to both test methods:

• loginWithIncorrectAndThenCorrectPassword
• loginWithDeprecatedSSLOptions

This matches the pattern already used in PulseSecurityIntegrationTest.

Testing

• Both tests pass locally (27.7s total, 0 failures)
• Code compiles and passes tests
• No timing-dependent failures observed

Impact

• Eliminates flaky test behavior in CI
• Improves test reliability and developer experience
• No functional changes to production code

For all changes, please confirm:

• Is there a JIRA ticket associated with this PR? Is it referenced in the commit message?
• Has your PR been rebased against the latest commit within the target branch (typically develop)?
• Is your initial contribution a single, squashed commit?
• Does gradlew build run cleanly?
• Have you written or updated unit tests to verify your changes?
• If adding new dependencies to the code, are these dependencies licensed in a way that is compatible for inclusion under ASF 2.0?