Migrate to rsbuild and swc

[RichDom2185]

• Closes Migrate to Vite #2933.
• Closes Clean up codebase post-migration to Vite/React19 #2990.
• Supersedes Migrate to Vite #2996.

---

Description

CRA is dead. Switch to rsbuild (which uses rspack → drop-in for webpack → powers CRA), but written in Rust so much faster.

• Switch to rsbuild for performance and maintenance
• Upgrade jest from v27 to v29
• Switch to swc parser (also happens to be written in rust lol) for jest for performance
• Remove CRA and CRACO

Type of change

• Code quality improvements

How to test

Everything should (and does) still work since it's effectively drop in.

Checklist

• I have tested this code
• I have updated the documentation

[coveralls]

Pull Request Test Coverage Report for Build 15025193352

Details

• 22 of 39 (56.41%) changed or added relevant lines in 5 files are covered.
• 7962 unchanged lines in 470 files lost coverage.
• Overall coverage increased (+14.0%) to 45.029%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
jsdom-env.js	0	4	0.0%
jest.config.js	0	6	0.0%
rsbuild.config.ts	0	7	0.0%

Files with Coverage Reduction	New Missed Lines	%
src/assets/BrickSvg.tsx	1	83.33%
src/commons/application/actions/CommonsActions.ts	1	91.67%
src/commons/application/reducers/CommonsReducer.ts	1	88.89%
src/commons/application/reducers/VscodeReducer.ts	1	88.89%
src/commons/application/tests/snapshots/Application.tsx.snap	1	0.0%
src/commons/assessment/AssessmentNotFound.tsx	1	85.71%
src/commons/collabEditing/CollabEditingActions.ts	1	92.86%
src/commons/controlBar/ControlBarCloseButton.tsx	1	92.86%
src/commons/controlBar/ControlBarStepLimit.tsx	1	75.0%
src/commons/controlBar/github/ControlBarGitHubButtons.tsx	1	83.02%

Totals
Change from base Build 14851045635:	14.0%
Covered Lines:	10139
Relevant Lines:	21111

---

💛 - Coveralls

[sayomaki]

Did a preliminary quick review of the code changes, although I'm not too sure for the testing updates. As long as they are passing (which they appear to do so), I think it should not be too far off.

I have also yet to run the code locally, so here are just some questions and ideas from me after glancing over the changes, let me know what you think!

[leeyi45]

Previously with Webpack it would compile away the dynamic import used to import modules (because there was no way to webpack-ignore something in js-slang). Should I check if js-slang can be switched away from using the function constructor, or should we leave things as is?

[RichDom2185]

Previously with Webpack it would compile away the dynamic import used to import modules (because there was no way to webpack-ignore something in js-slang). Should I check if js-slang can be switched away from using the function constructor, or should we leave things as is?

@leeyi45 I think I saw this comment somewhere, but wasn't too sure what it entailed in terms of testing.

I had tried running some module code in the playground and it seems to work, though I'm not sure if that is what you mean.

Would you mind checking as well? Thank you!

[leeyi45]

Previously with Webpack it would compile away the dynamic import used to import modules (because there was no way to webpack-ignore something in js-slang). Should I check if js-slang can be switched away from using the function constructor, or should we leave things as is?

@leeyi45 I think I saw this comment somewhere, but wasn't too sure what it entailed in terms of testing.

I had tried running some module code in the playground and it seems to work, though I'm not sure if that is what you mean.

Would you mind checking as well? Thank you!

Did a quick review. I think on the js-slang side we should move toward building for ESM rather than CommonJS, which would make this change easier. Either ways, I've put the ignore warnings config back in for the dynamic module import