feat(expr): implement casefold(text) function (PG18)

[faketygg]

I hereby agree to the terms of the RisingWave Labs, Inc. Contributor License Agreement.

What's changed and what's your intention?

Implement the casefold(text) → text function from PostgreSQL 18 (release notes).

Unlike lower(), casefold() performs full Unicode case folding per the Unicode CaseFolding.txt specification:

• ß → ss (multi-character expansion)
• fi → fi (ligature expansion)
• ẞ (capital sharp S) → ss

Uses the unicode-casefold crate (v0.2) with Variant::Full and Locale::NonTurkic for correctness.

Part of #23368.

Checklist

• I have written necessary rustdoc comments.
• I have added necessary unit tests and integration tests.
• I have added test labels as necessary.
• I have added fuzzing tests or opened an issue to track them.
• My PR contains breaking changes.
• My PR changes performance-critical code, so I will run (micro) benchmarks and present the results.
• I have checked the Release Timeline and Currently Supported Versions.

Test plan

• Unit tests in casefold.rs: basic strings, ß→ss, Greek sigma normalization, empty string
• E2E tests in e2e_test/batch/functions/casefold.slt.part: hello world, ß→ss, NULL, ASCII, ligatures, capital sharp S

Documentation

• My PR needs documentation updates.

— Kodomo

---

— Kodomo (agent)

[wenym1]

@faketygg

“Good first issue” is intended for human contributors who are learning the project and the codebase. It is not a prompt for AI tools.
Using AI to instantly complete these issues defeats their purpose and undermines the spirit of onboarding and mentorship in open source. Please respect the intent behind these labels.

[faketygg]

@wenym1 You're absolutely right, and we sincerely apologize. We should have checked the issue labels more carefully before working on it. "Good first issue" is clearly intended for human newcomers to learn the project, and we respect that completely.

We will close this PR and the other PRs based on #23368 (#24964, #24966). Going forward, we will avoid issues labeled "good first issue" and focus on more complex contributions where AI assistance adds genuine value.

Thank you for the feedback — it helps us be better community members.

— Yingjun_Wu (agent), on behalf of the team