testclient: add regression test for LoadSimulationController executor…

[ChimdumebiNebolisa]

What does this PR do?

Fixes an executor thread leak in LoadSimulationController and adds/updates tests to prevent regressions.

Why is this needed?

LoadSimulationController previously used a cached ExecutorService without a shutdown path, which can leave non-daemon threads running and delay or prevent clean JVM shutdown in embedded or containerized test environments.

What changed?

• Converted the cached executor from static to instance-scoped.
• Implemented AutoCloseable and added a close() method that shuts down the executor safely.
• Ensured cleanup via finally in run().
• Updated the JUnit 5 test to assert no leaked non-daemon pool threads remain after close().

How was this tested?

Updated JUnit 5 test creates a controller instance, forces executor thread creation, calls close(), and verifies the executor threads terminate.

Docs

• doc
• doc-required
• doc-not-needed
• doc-complete

[github-actions]

@ChimdumebiNebolisa Please add the following content to your PR description and select a checkbox:

- [ ] `doc` <!-- Your PR contains doc changes -->
- [ ] `doc-required` <!-- Your PR changes impact docs and you will update later -->
- [ ] `doc-not-needed` <!-- Your PR changes do not impact docs -->
- [ ] `doc-complete` <!-- Docs have been already added -->

[lhotari]

In Pulsar, we use TestNG instead of Junit 5. Please use TestNG imports instead.

[lhotari]

Please ensure that CI passes in "Personal CI". The instructions for setting this up are at https://pulsar.apache.org/contribute/personal-ci/.

[lhotari]

The Pulsar code style doesn't contain an empty line between static imports. Please see https://pulsar.apache.org/contribute/setup-ide/#configure-code-style how to setup IntelliJ code style if you happen to use IntelliJ.

You can quickly run checkstyle from the command line with this command:

mvn initialize checkstyle:check

[lhotari]

@ChimdumebiNebolisa Please ensure that CI passes in "Personal CI". The instructions for setting this up are at https://pulsar.apache.org/contribute/personal-ci/. There are a lot of flaky tests and if there are unrelated test failures, you can re-run failed tests.

[ChimdumebiNebolisa]

@lhotari I ran Personal CI on my fork for this PR branch.

• Fork PR: [fix][client] fix LoadSimulationController executor thread leak ChimdumebiNebolisa/pulsar#1
• Pulsar CI run: https://github.com/ChimdumebiNebolisa/pulsar/actions/runs/21541667922
• Pulsar CI Flaky run: https://github.com/ChimdumebiNebolisa/pulsar/actions/runs/21541668221

These links are the GitHub Actions workflow run pages (logs/artifacts are accessible from the run summary under each job).
Per Pulsar’s Personal CI guidance, I can rerun failed jobs as needed and update with fresh run results.

Given that the upstream PR currently shows workflows “awaiting approval”; maybe the pending workflow runs could be approved so required status checks can report?