Fix UTF-32 surrogate handling in toWellFormed

[Copilot]

The toWellFormed function in fallback/utf32.js incorrectly preserved consecutive surrogate codepoints when they formed what would be a valid pair in UTF-16. UTF-32 does not use surrogate pairs—all values in the surrogate range (0xD800-0xDFFF) are invalid and must be replaced individually.

Changes

• fallback/utf32.js: Simplified surrogate handling to unconditionally replace all codepoints in 0xD800-0xDFFF range with replacement character (0xFFFD)
• Removed pair-checking logic that was incorrectly ported from UTF-16 implementation

Example

// Before: consecutive surrogates incorrectly preserved
const input = new Uint32Array([0x41, 0xD800, 0xDC00, 0x42])
toWellFormed(input)  // [0x41, 0xD800, 0xDC00, 0x42] ❌

// After: all surrogates replaced as required by UTF-32
toWellFormed(input)  // [0x41, 0xFFFD, 0xFFFD, 0x42] ✓

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Review the codebase and identify mistakes</issue_title>
<issue_description>Ignore tests/wpt and other vendored tests

Note: wif using synchronous path for async implementation is not an issue, it's done for perf, as underlying method is always sync (again, for perf).

Pay close attention to corner cases</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Review the codebase and identify mistakes #43

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[ChALkeR]

Ok this makes sense but will do that separately
UTF-32 is not shipped yet and was not carefully tested