fix: preserve reexport call context in concatenation

[LingyuCoder]

Summary

• Preserve the consuming import expression's call and ASI semantics while faster module concatenation follows a structured direct re-export.
• Strengthen the existing CommonJS interop case with a strict, this-sensitive default export.

Faster module concatenation previously used the re-export edge's fixed call and asi_safe metadata when resolving the final binding. This discarded the consuming import's context and could render a CommonJS default call as getter.a(...) instead of getter()(...), changing the function's this value.

After this fix, enabling or disabling experiments.fasterModuleConcatenation produces byte-for-byte identical output across all five projects in rstackjs/build-tools-performance (react-1k, react-5k, react-10k, popular-libs, and ui-components).

Related links

• Follow-up to perf: avoid parsing concatenated module codegen #14852

Validation

• pnpm run build:binding:dev
• pnpm run test:unit (9208 passed, 4 skipped)
• cargo fmt --all --check
• Cold-cache faster on/off output comparison for all rstackjs/build-tools-performance projects; the full output tree is byte-for-byte identical

Checklist

• Tests updated (or not required).
• Documentation updated (or not required).

[github-actions]

📦 Binary Size-limit

Comparing c17be40 to chore(vscode): use Rstack extension as default formatter (#15245) by Jiahan Chen

🙈 Size remains the same at 68.11MB

[github-actions]

Rsdoctor Bundle Diff Analysis

Found 5 projects in monorepo, 0 projects with changes.

📊 Quick Summary

Project	Total Size	Gzip Size	Change	Gzip Change
popular-libs	1.7 MB	556.5 KB	0	0
react-10k	5.6 MB	1.3 MB	0	0
react-1k	823.1 KB	218.4 KB	0	0
react-5k	2.7 MB	669.3 KB	0	0
ui-components	4.9 MB	1.4 MB	0	0

Generated by Rsdoctor GitHub Action

[codspeed]

Merging this PR will not alter performance

✅ 50 untouched benchmarks
⏩ 47 skipped benchmarks¹

---

_{Comparing seal/faster-concat-output-parity (c17be40) with main (77f4caa)}

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. ↩