[FLINK-30751] [docs] Remove references to disableDataSync in RocksDB documentation

[dchristle]

What is the purpose of the change

This PR corrects and improves the readability of the documentation for RocksDB predefined option sets. References to disableDataSync, which is not set and has been removed from RocksDB for some time, are removed.

While fixing this error, I also noticed other confusing aspects of this documentation. The options are not sorted, which makes it harder for a reader to find a particular one, and for maintainers to ensure the code & documentation match. I found an error of this kind in SPINNING_DISK_OPTIMIZED_HIGH_MEM, where Bloom filter usage is enabled but not documented. The unsorted documentation/code may have played a role in this accidental omission. I kept the Bloom filter setting, so that users' job performance does not change unexpectedly between versions, and I added the Bloom filter setting to the documentation along with a description of where its input parameters are set.

References to Flink not needing to sync changes to the file system due to its checkpointing ability are also present, but it is confusing to a reader because this point isn't relevant to any of the listed options in the Java documentation. setUseFsync(false) is documented for each option set in the Python documentation, but isn't in the Java documentation. setUseFsync(false) is not set in the predefined options, so we should not emphasize it. Instead, I added a passing reference to it at the top of both Java and Python introductions.

Brief change log

• Remove references to disableDataSync in RocksDB documentation. This is not set, and has been removed from RocksDB for some time.
• Sort options in both Java and Python documentation. Sort the code setting configuration to match the sorted-order documentation.
• Make Python documentation consistent with Java documentation by removing setUseFsync references on each option.
• Change use of setIncreaseParallelism in Python documentation to setMaxBackgroundJobs, which matches the Java usage.
• Add missing Bloom filter option to the SPINNING_DISK_OPTIMIZED_HIGH_MEM documentation, and describe where its parameters are set.

Verifying this change

This change is a trivial rework / code cleanup without any test coverage.

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

[flinkbot]

CI report:

• 38f4da5 UNKNOWN
• f557045 Azure: SUCCESS

Bot commands

The @flinkbot bot supports the following commands:

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

[Myasuka]

Thanks for the contribution!

I think this PR should be split into two commits:

1. Remove references to disableDataSync in RocksDB documentation.
2. Align pyflink rocksdb options with java side.

[dchristle]

Hi @Myasuka,

Thanks for reviewing this PR! I split the changes into separate commits so that each change is separated & easier to review. The first commit removes disableDataSync, and the last commit aligns the PyFlink RocksDB documentation with its Java equivalent, as you requested.

[dchristle]

Hi @Myasuka Is there any chance we can merge this PR?

[Myasuka]

@dchristle Thanks for the reminder, I will take a look recently.

[Myasuka]

LGTM, @dchristle could you rebase it with latest master branch so that we can get a green CI.

[dchristle]

@Myasuka Got it - I just rebased on the latest master branch.

[dchristle]

@Myasuka Thanks for your review & help here!