[Docs][Connector-V2] Improve OceanBase connector docs

[DanielCarter-stack]

Summary

• Improve OceanBase source and sink documentation in English and Chinese.
• Add missing source options for table path, table list, split tuning, regex matching, and type narrowing.
• Add missing sink options for generated SQL, save modes, XA exactly-once, timed flushing, CDC/upsert, and multi-table writes.
• Add OceanBase MySQL and Oracle compatible mode examples for source and sink usage.

Verification

• Ran git diff --check.
• Ran a markdown table column-count check for the updated OceanBase docs.
• Checked recent PRs mentioning OceanBase docs to avoid duplicating another docs PR.

[DanielLeens]

Thanks for working on this. I re-reviewed the latest head 3ec8ffb856f1 from scratch.

What this PR fixes

• User pain: the OceanBase source/sink pages had fallen behind the current JDBC implementation, especially around exactly-once, timer-based flushing, multi-table read/write, table_list / table_path, save modes, and the mysql-vs-oracle compatibility split.
• Fix approach: update both the EN and ZH OceanBase source/sink docs so they describe the current JDBC-backed capability surface much more completely.
• One-line summary: after tracing the new claims back to the real JDBC source/sink code paths, I do not see a new accuracy blocker in the current version.

Full call path I checked

OceanBase source docs
  -> source { Jdbc { ... compatible_mode = "mysql"/"oracle" ... } }
      -> JdbcSourceFactory.optionRule()
      -> JdbcSourceConfig.of()
      -> JdbcSourceTableConfig.of()
      -> JdbcSource
          -> JdbcCatalogUtils.getTables(...)

OceanBase sink docs
  -> sink { jdbc { ... } }
      -> JdbcSinkFactory.optionRule()
          -> BATCH_INTERVAL_MS
          -> IS_EXACTLY_ONCE / XA_DATA_SOURCE_CLASS_NAME
          -> MULTI_TABLE_SINK_REPLICA
          -> schema/data save mode
      -> JdbcSink
          -> SupportMultiTableSink
      -> JdbcExactlyOnceSinkWriter (XA path)

That is the important framing here: these OceanBase pages are really documenting the OceanBase usage profile of the Jdbc connector, so the correctness check is whether the new option/feature claims actually exist on the JDBC source/sink path. The main ones added in this PR do line up with the current code path.

Findings

I did not find a new blocker in the current version.

The bigger doc claims I explicitly cross-checked were:

• source-side table_path / table_list / where_condition / split options
• sink-side batch_interval_ms
• sink-side exactly-once via is_exactly_once + xa_data_source_class_name
• multi-table write/read claims
• mysql-compatible vs oracle-compatible mode guidance

Those all map back to real JDBC source/sink options or interfaces in the current codebase, so I’m comfortable with the updated docs.

Review conclusion

Conclusion: can merge

1. Blocking items

• None from my side.

2. Non-blocking suggestions

• None that I would hold the merge for in this round.

Overall this is a good docs catch-up pass for OceanBase, and it is much closer to the real capability surface now. Happy to approve.

[DanielLeens]

Rechecked the latest head: the source-level blocker from my previous Daniel round is still closed for me, and the current Build gate is now green on this unchanged head.

At this point I do not see a new code-side blocker from Daniel on this revision. For queue bookkeeping, this branch is still diverged against the latest dev (ahead_by=1, behind_by=12), but because the required Build signal is green now, I am not asking for a sync-only retry from Daniel's side here.

If the repository still needs a write-capable maintainer to refresh the formal review state or dismiss an older visible changes requested entry, that final gate may need to be handled separately from Daniel's source-level recheck.

[SEZ9]

Thanks for working on this. I re-reviewed the latest head 3ec8ffb856f1 from scratch.

Review conclusion

Conclusion: LGTM, can merge

Nice work — thanks again for the contribution. Happy to discuss any of the above.

[DanielLeens]

Thanks @SEZ9 for the rereview. I rechecked the same unchanged head and I also do not see a reopened Daniel-side source blocker here.