[FLINK-29053] Hybrid shuffle has concurrent modification of buffer when compression is enabled

[reswqa]

What is the purpose of the change

When the downstream thread obtains the buffer and consuming it, if the data is compressed in the spilling thread and copied to the original buffer in the same time, since the two threads share the same memory data, the consuming thread will consume incorrect data, causing problems such as deserialize the data disorder. This pr will fix this problem.

Brief change log

• Move the compression operation to the write thread and store the compressed buffer directly in memory.

Verifying this change

This change is already covered by existing tests

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

[flinkbot]

CI report:

• e6a8a62 Azure: SUCCESS

Bot commands

The @flinkbot bot supports the following commands:

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

[reswqa]

@flinkbot run azure

[xintongsong]

@flinkbot run azure

[xintongsong]

LGTM