perf: improve bundle splitting

[LingyuCoder]

Summary

• If the chunk filter is not a JavaScript binding function, handle it synchronously to reduce the time consumption introduced by "join_all".
• Add disconnect_chunks_and_modules and connect_chunk_and_modules on chunk graph to reduce the number of times of obtaining chunk_graph_chunk and chunk_graph_module.
• Call get_module_chunks in parallel to reduce the number of calls in the subsequent prepare_module_group_map.
• Cache the results of get_combs based on used_exports to prevent repeated calculations for the same module.
• When removing duplicate modules, traverse the chunk group with a smaller number of modules to reduce the hashing cost.
• When the cache group name exists, it is quite likely that the same module will obtain the same cache group key under different chunk sets. In this case, there is no need to add the module again.

Since the running time of bundle splitting is unstable, it is currently judged to be caused by a large number of invocations of rayon for concurrent module removal. There are approximately 5,000 invocations. At this time, if the CPU is busy, performance degradation may occur due to the invocations being blocked. Therefore, the following time consumption is for reference only.

Before:

After:

Related links

Checklist

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

[netlify]

✅ Deploy Preview for rspack ready!

Name	Link
🔨 Latest commit	970baf2
🔍 Latest deploy log	https://app.netlify.com/projects/rspack/deploys/689b3e2d8f816c00081bc860
😎 Deploy Preview	https://deploy-preview-11364--rspack.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[github-actions]

📦 Binary Size-limit

Comparing 970baf2 to refactor: inline const parser plugin (#11359) by Gengkun

❌ Size increased by 16.50KB from 47.56MB to 47.58MB (⬆️0.03%)

[codspeed-hq]

CodSpeed Performance Report

Merging #11364 will not alter performance

_{Comparing perf/bundle-splitting (970baf2) with main (19b2442)}

🎉 Hooray! codspeed-node just leveled up to 4.0.1!

A heads-up, this is a breaking change and it might affect your current performance baseline a bit. But here's the exciting part - it's packed with new, cool features and promises improved result stability 🥳!
Curious about what's new? Visit our releases page to delve into all the awesome details about this new version.

Summary

✅ 17 untouched benchmarks

[Copilot]

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.