[FLINK-39962] [runtime] Fix flaky DeclarativeSlotPoolBridgeTest

[qiuyanjun888]

What is the purpose of the change

This pull request fixes the flaky DeclarativeSlotPoolBridgeTest.testAcceptingOfferedSlotsWithoutResourceManagerConnected reported in FLINK-39962.

The Jira describes a test-only threading race: the test used the regular forMainThread() test executor, whose scheduled timeout callbacks can run concurrently with the test thread while close() copies pendingRequests, intermittently producing NegativeArraySizeException.

Brief change log

• Use a ManuallyTriggeredScheduledExecutorService for the affected test.
• Manually complete the deferred slot-request declaration task when slotRequestMaxInterval is positive.
• Keep production slot-pool code unchanged and avoid triggering unrelated request/idle/batch timeout tasks in this test.

Verifying this change

This change is covered by the existing affected test and the containing test class:

• ./mvnw -pl flink-runtime -Dtest=DeclarativeSlotPoolBridgeTest#testAcceptingOfferedSlotsWithoutResourceManagerConnected -DfailIfNoTests=false -DskipITs -Dfast -Drat.skip=true -Dcheckstyle.skip=true -Dspotless.check.skip=true test
• ./mvnw -pl flink-runtime -Dtest=DeclarativeSlotPoolBridgeTest -DfailIfNoTests=false -DskipITs -Dfast -Drat.skip=true -Dcheckstyle.skip=true -Dspotless.check.skip=true test
• ./mvnw -pl flink-runtime -DskipTests -DskipITs -Drat.skip=true spotless:check
• ./mvnw -pl flink-runtime -DskipTests -DskipITs -Drat.skip=true checkstyle:check

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

• Dependencies (does it add or upgrade a dependency): no
• The public API, i.e., is any changed class annotated with @Public(Evolving): no
• The serializers: no
• The runtime per-record code paths (performance sensitive): no
• Anything that affects deployment or recovery: JobManager (and its components), Checkpointing, Kubernetes/Yarn, ZooKeeper: no
• The S3 file system connector: no

Documentation

• Does this pull request introduce a new feature? no
• If yes, how is the feature documented? not applicable

---

Was generative AI tooling used to co-author this PR?

• Yes (please specify the tool below)

Generated-by: Hermes Agent (OpenAI GPT-5.5)

[flinkbot]

CI report:

• 26ee032 Azure: FAILURE

Bot commands

The @flinkbot bot supports the following commands:

• @flinkbot run azure re-run the last Azure build

[spuru9]

LGTM.
Can you work on making build green.

[qiuyanjun888]

@flinkbot run azure

[qiuyanjun888]

This PR is focused on runtime / slotpool flaky test.
There has already been community review/comment, but it still needs maintainer review.

@RocMarshal @1996fanrui could you please take a look when you have time and advise whether this approach is acceptable for this area and can move forward?

[davidradl]

I am curious can this race occur in the main code ?