[branch-4.1][fix](fe) Disable expression ZoneMap filtering by default

[mrhhsg]

What problem does this PR solve?

Issue Number: None

Related PR: #65263

Problem Summary:

Expression ZoneMap filtering was enabled by default when it was introduced on branch-4.1. This change disables enable_expr_zonemap_filter for regular new sessions and enables it in fuzzy mode when Config.pull_request_id % 2 == 0, so the fuzzy setting is deterministic for each PR while covering both enabled and disabled paths across PRs.

The dedicated expression ZoneMap regression suites now enable the session variable explicitly, keeping their profile-counter assertions deterministic regardless of the global default or fuzzy-mode PR parity.

For mixed-version compatibility, the absent Thrift-field behavior remains legacy-enabled. New FE sessions always serialize the session value explicitly through SessionVariable.toThrift().

Release note

Disable expression ZoneMap filtering by default on branch-4.1 and enable it for fuzzy-mode sessions with even pull request IDs.

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: the fuzzy value follows the existing PR-parity assignment pattern.

Validation:

• git diff --check passed.

• Behavior changed:

  • No.
  • Yes. Expression ZoneMap filtering is disabled by default for new sessions and enabled in fuzzy mode for even pull request IDs.

• 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?

[mrhhsg]

run buildall

[hello-stephen]

FE UT Coverage Report

Increment line coverage 0.00% (0/1) 🎉
Increment coverage report
Complete coverage report

[hello-stephen]

FE Regression Coverage Report

Increment line coverage 0.00% (0/23) 🎉
Increment coverage report
Complete coverage report

[mrhhsg]

/review

[github-actions]

Reviewed the default flip for enable_expr_zonemap_filter, the fuzzy-mode PR-parity assignment, and the two regression-suite opt-ins. I did not find a correctness issue worth an inline comment.

Critical checkpoints:

• Existing review context: no prior inline comments or replies in the prepared review bundle.
• Session/default behavior: new FE-created sessions serialize the explicit false default through SessionVariable.toThrift(), while the BE absent-thrift-field path remains legacy-enabled through expr_zonemap::is_expr_zonemap_filter_enabled().
• Scan gating: OLAP segment/page pruning and Parquet row-group/page pruning all use the centralized BE helper.
• Regression coverage: both changed expression-ZoneMap suites explicitly enable enable_expr_zonemap_filter and keep the required common-expression/profile settings before their profile-counter assertions.
• Fuzzy mode: the PR-id parity assignment is deterministic; non-PR fuzzy mode with pull_request_id = 0 enables the feature, which I do not see as a regression in this PR.
• User focus: no additional user-provided focus was present.
• Review instruction note: the repository-local code-review skill required by the prompt was not discoverable in this checkout, so I followed the explicit prompt and ledger workflow.

Validation was static only; I did not run builds or regression tests in this review runner.