[fix][broker] Fix ManagedCursorImpl.asyncDelete() method may lose previous async mark delete properties in race condition

[oneby-wang]

Motivation

ManagedCursorImpl.asyncDelete() method may lose previous async mark delete properties in race condition.

Modifications

1. Add testAsyncDeleteNeverLoseMarkDeleteProperties() test to reproduce the issue.
2. Pass null to internalAsyncMarkDelete() method's properties param in asyncDelete() method, which depends on [fix][broker] Fix markDeletedPosition race condition in ManagedLedgerImpl.maybeUpdateCursorBeforeTrimmingConsumedLedger() method #25110.
3. Run test to verify the code change.

Verifying this change

• Make sure that the change passes the CI checks.

Does this pull request potentially affect one of the following parts:

If the box was checked, please highlight the changes

• Dependencies (add or upgrade a dependency)
• The public API
• The schema
• The default values of configurations
• The threading model
• The binary protocol
• The REST endpoints
• The admin CLI options
• The metrics
• Anything that affects deployment

Documentation

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

Matching PR in forked repository

PR in forked repository: oneby-wang#25

[lhotari]

Depends on #25110

[lhotari]

LGTM. Good catch @oneby-wang . Please update the description to contain the detail that this PR depends on #25110.

[oneby-wang]

@lhotari PR description updated.

Some more places need further working:

1. ManagedLedgerImpl.advanceCursorsIfNecessary()
2. PersistentMessageExpiryMonitor.findEntryComplete()
3. ManagedCursorImpl.internalResetCursor()

I will address the three above references in separate PRs. First, I will write test cases to reproduce the issue, then fix it using the approach referenced in this PR.

Please kick me if I miss something.

[lhotari]

/pulsarbot rerun-failure-checks

[oneby-wang]

ExtensibleLoadManagerImplTest in the flaky test suite always fails no matter how many times I rerun it in my personal CI(with or without this PR).

https://github.com/oneby-wang/pulsar/actions/runs/21156674861/job/60853773530?pr=25

https://github.com/oneby-wang/pulsar/actions/runs/21158046247/job/60853358807?pr=11

Seems some code changes break this flaky test.