test(css): enable webpack css module cases

[intellild]

Summary

• Port webpack's native CSS modules coverage into configCases/css/css-modules and store large CSS outputs as case-local file snapshots.
• Remove the remaining CSS-related test.filter.js files so the CSS config, serial, hot, diagnostics, and extract-css cases run by default.
• Link css-module-lexer to the remote git revision used for the new parser behavior, then fix the surfaced compatibility gaps in CSS Modules parsing, extract-css rebuilds, and lightningcss warning diagnostics.

Examples

• A CSS module can now import ICSS values from another local CSS file, for example @value primary from "./colors.module.css", while still resolving composes and generated reexports consistently.
• A selector written without whitespace before a CSS Modules mode marker, for example .foo:local(.bar), is preserved with a warning instead of breaking local ident generation in plain CSS files.
• Rebuilding an extracted CSS module after an imported module changes now uses a content-sensitive CSS dependency identity, so a case that changes imported CSS from red to green no longer reuses stale extracted CSS.
• Invalid lightningcss input now surfaces warnings in stats/raw-warning output instead of being swallowed by the loader or minimizer path.

Validation

• pnpm run build:binding:dev
• pnpm run test:unit
• pnpm run test:rs
• cargo clippy --workspace --all-targets --locked -- -D warnings
• cargo fmt --all -- --check
• git diff --cached --check

---

by OpenAI Codex

[github-actions]

📦 Binary Size-limit

Comparing 0433047 to perf(hook): reduce hook macro expansion size and optimize performance (#14633) by AsyncIter

❌ Size increased by 40.00KB from 66.38MB to 66.42MB (⬆️0.06%)

[codspeed-hq]

Merging this PR will not alter performance

✅ 43 untouched benchmarks
⏩ 47 skipped benchmarks¹

---

_{Comparing codex/css-module-tests (0433047) with main (f489568)}

Footnotes

1. 47 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩