fix(module/source.rs,namespace.rs,synthetic.rs): convert panics to EngineError::Panic using js_expect

[KaustubhOG]

Part of #3241. It changes the following:

core/engine/src/module/source.rs: converted 24 panics to use js_expect.
core/engine/src/module/namespace.rs: converted 11 panics to use js_expect.
core/engine/src/module/synthetic.rs: converted 4 panics to use js_expect.

This completes all real panics in module/. The remaining .expect() calls in module/ are either in non-JsResult functions (load(), inner_load(), load_link_evaluate(), link(), execute_async()), inside debug_assert! blocks, or in #[cfg(test)] blocks and are left as-is.

[github-actions]

Test262 conformance changes

Test result	main count	PR count	difference
Total	52,963	52,963	0
Passed	49,935	49,935	0
Ignored	2,207	2,207	0
Failed	821	821	0
Panics	0	0	0
Conformance	94.28%	94.28%	0.00%

Tested main commit: 5acb570038007c606d75b17bed6ee9a02117ce97
Tested PR commit: a2a875b326340ee89202991d19340cee4c15393d
Compare commits: 5acb570...a2a875b

[codecov]

Codecov Report

❌ Patch coverage is 50.00000% with 9 lines in your changes missing coverage. Please review.
✅ Project coverage is 58.98%. Comparing base (6ddc2b4) to head (a2a875b).
⚠️ Report is 853 commits behind head on main.

Files with missing lines	Patch %	Lines
core/engine/src/module/source.rs	46.66%	8 Missing ⚠️
core/engine/src/module/namespace.rs	0.00%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #5064       +/-   ##
===========================================
+ Coverage   47.24%   58.98%   +11.73%     
===========================================
  Files         476      563       +87     
  Lines       46892    62569    +15677     
===========================================
+ Hits        22154    36905    +14751     
- Misses      24738    25664      +926     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[KaustubhOG]

[KaustubhOG]

Remaining module/ panics audit — all invariants, zero convertible

File(s)	Hits	Reason
module/source.rs	5	.expect("should be linking") — state machine invariant, transition only reachable from correct state. .expect("cannot fail for the %Promise% intrinsic") / .expect("%Promise% constructor must always return a Promise object") / .expect("async modules cannot directly throw") — spec-guaranteed %Promise% intrinsic invariants. panic! on exhaustive match arm — unreachable by construction.
module/synthetic.rs	2	.expect("default resolve functions cannot throw and must return a promise") — spec-guaranteed promise invariant. .expect("should have the module environment") — environment set during link phase, guaranteed present at evaluate.
module/namespace.rs	2	Both .expect("6. Assert: binding is a ResolvedBinding Record.") — verbatim spec assert, binding is guaranteed non-empty by the caller's resolve step.
module/loader/embedded.rs	1	.expect("Invalid compress type") — called on a compile-time &'static str constant from the macro, cannot be an invalid value at runtime.
module/mod.rs (tests)	7	All inside #[test] functions — .unwrap() on context builder and module parse. Not in scope for #3241.
Total	17	0 convertible

All remaining panics fall into: spec-mandated assert invariants, module state machine invariants, %Promise% intrinsic guarantees, compile-time constant .expect, or test assertions. None are suitable for EngineError conversion.

[jedel1043]

Uh I'd argue these kind of state transition steps should also be engine panics, since those have the highest probability of being incorrect

Also, I think I've mentioned the spec might also have bugs, so it is safer to engine panic on spec assertions.

[KaustubhOG]

Uh I'd argue these kind of state transition steps should also be engine panics, since those have the highest probability of being incorrect

Also, I think I've mentioned the spec might also have bugs, so it is safer to engine panic on spec assertions.

Yup , just saw yours comment on the 5063 , will convert the statetranstions and spec assertions ones .