[hotfix][docs] Remove dangling references to removed Hive connector docs

[hulincup]

What & Why

The Hive connector was removed from master in FLINK-33786 (commit 0b99f169, 2025-02-17). That change deleted the entire docs/content/docs/connectors/table/hive/ directory, but several other documents still contained {{< ref >}} shortcodes pointing at the removed pages, leaving dangling references.

This PR cleans up those dangling references.

Changes

• docs/connectors/table/overview.md (EN + ZH): Removed the obsolete Apache Hive entry from the connectors table. The connector no longer exists in master, and both its name link and the Supported Versions link pointed to the deleted hive/overview page.
• sql/hive-compatibility/hive-dialect/overview.md (EN + ZH), sql/catalogs.md (EN + ZH), dev/table/modules.md (EN + ZH), sql/reference/utility/jar.md (EN + ZH), sql/interfaces/sql-gateway/hiveserver2.md (EN) / sql/hive-compatibility/hiveserver2.md (ZH), concepts/sql-table-concepts/versioned_tables.md (ZH): Replaced the broken {{< ref "docs/connectors/table/hive/..." >}} links with plain text (the original link text is retained, only the dead link markup is removed). Valid links to other connectors (e.g. JDBC, HBase) and other pages are left untouched.

Scope

13 files in total (English + Chinese), purely documentation. No code changes. All edits only remove dead ref links pointing at the removed connectors/table/hive/ directory; no content is rewritten beyond dropping the link markup.

This is a docs-only hotfix following up on FLINK-33786; no JIRA ticket was created for the cleanup itself.

🤖 Generated with Claude Code

[flinkbot]

CI report:

• 32fe01f Azure: SUCCESS

Bot commands

The @flinkbot bot supports the following commands:

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

[hulincup]

CI failure analysis

The Azure CI run (build 77606) failed in the Compile task (mvn clean deploy ... -DskipTests -Pcheck-convergence, ~50 min run, Bash exited with code 1).

This PR is docs-only (13 .md files under docs/content/), which does not participate in the Maven compile/deploy phase. The failure appears unrelated to this change — likely a flaky run (recent merged PRs #28846 / #28802 / #28844 all passed Azure CI green).

Re-triggering to confirm: @flinkbot run azure

[hulincup]

@flinkbot run azure

[hulincup]

Friendly ping — the Azure CI failure appears to be a flaky run (docs-only change, Maven compile does not read these files; recent merged PRs are green). I tried @flinkbot run azure but it did not trigger a new build (likely a permission issue for fork contributors).

Could a committer help re-trigger the Azure CI (@flinkbot run azure)? Thanks!

[hulincup]

Root cause of the CI failure identified

I downloaded the full Azure log and found the exact failure. It is not related to this docs-only change.

Failing module: flink-runtime-web (the Angular web dashboard, not docs/)

Failing step: frontend-maven-plugin:npm (npm install) → npm ci in flink-runtime-web/web-dashboard

Exact error:

[INFO] Running 'npm ci ${npm.proxy}' in /__w/2/s/flink-runtime-web/web-dashboard
[INFO] npm error code ECONNRESET
[INFO] npm error network aborted
[INFO] npm error network This is a problem related to network connectivity.
[INFO] npm error network In most cases you are behind a proxy or have bad network settings.

This is an infrastructure flake — the Azure runner's npm ci was aborted by a transient network reset (ECONNRESET) while fetching npm packages. It has nothing to do with the 13 .md files this PR touches under docs/content/.

The earlier @flinkbot run azure did not actually spawn a new build (the bot kept referencing build 77606). Pushing an empty commit to retrigger a fresh CI run.