[feature](compute group) Support show compute groups in non cloud mode (#66697)

[deardeng]

pick from #66697

Support SHOW CLUSTERS and SHOW COMPUTE GROUPS in non-cloud mode by exposing authorized resource groups and their backend counts.

What problem does this PR solve?

Issue Number: close #xxx

Related PR: #xxx

Problem Summary:

Release note

None

Check List (For Author)

• Test

  • Regression test
  • Unit Test
  • Manual test (add detailed scripts or steps below)
  • No need to test or manual test. Explain why:
    • This is a refactor/code format and no logic has been changed.
    • Previous test can cover this change.
    • No code files have been changed.
    • Other reason

• Behavior changed:

  • No.
  • Yes.

• Does this need documentation?

  • No.
  • Yes.

Check List (For Reviewer who merge this PR)

• Confirm the release note
• Confirm test cases
• Confirm document
• Add branch pick label

[hello-stephen]

Thank you for your contribution to Apache Doris.
Don't know what should be done next? See How to process your PR.

Please clearly describe your PR:

1. What problem was fixed (it's best to include specific error reporting information). How it was fixed.
2. Which behaviors were modified. What was the previous behavior, what is it now, why was it modified, and what possible impacts might there be.
3. What features were added. Why was this function added?
4. Which code was refactored and why was this part of the code refactored?
5. Which functions were optimized and what is the difference before and after the optimization?

[deardeng]

run buildall

[hello-stephen]

FE Regression Coverage Report

Increment line coverage 23.53% (4/17) 🎉
Increment coverage report
Complete coverage report

[freemandealer]

/review

[github-actions]

Review result: approval; no blocking issue found.

Critical checkpoint conclusions:

• Goal and scope: The two-file change is focused and achieves the stated non-cloud behavior for both SHOW CLUSTERS and SHOW COMPUTE GROUPS: resource-group/location-tag names and backend counts are returned with the intended two-column metadata, while the cloud seven-column path remains unchanged.
• Data correctness and authorization: Non-cloud visibility uses the same connection-scoped compute-group membership as query scheduling. Default, single-tag, multi-tag, root/admin all-backend, and invalid-user cases were traced; no extra group exposure, dropped allowed group, or throwing subtype path survived review. Exact same-tag multi-backend counts and deterministic order are covered by the changed unit test.
• Compatibility and result delivery: Both SQL spellings continue through the same parser/builder path, mode-specific title casing and row widths align, and MySQL/Thrift delivery serializes the metadata and rows together. No persistence, FE/BE wire, storage-format, rolling-upgrade, or dynamic-configuration surface is introduced.
• Concurrency, lifecycle, and error handling: Backend membership is read from one immutable registry snapshot. Concurrent tag changes retain the existing administrative observation semantics shared by scheduling; no new lock, lifecycle transition, silent error, or durable inconsistent state was found.
• Tests and coverage: The changed FE test covers cloud and non-cloud metadata, exact counts for multiple backends sharing a tag, deterministic ordering, allowed-tag filtering, and a configured tag with no matching backend. Static mode changes are isolated by per-method setup and Surefire's non-reused test JVMs. No local build or test was run because the authoritative review instructions prohibit builds in this runner.
• Performance and observability: The non-cloud path is one linear backend scan plus sorting by the small set of group names, with no material hot-path, memory, or logging concern.
• User focus: No additional user-provided review focus was supplied.

Round 1 used two complementary full-review lanes and one separate risk-focused lane; all returned NO_NEW_VALUABLE_FINDINGS, and the main agent independently resolved every initial risk item with concrete code evidence.

[deardeng]

@freemandealer - -!