feat(cli): add --time flag to report script execution timing

[Nakshatra480]

Summary

This adds a --time flag to the Boa CLI that prints how long parsing and execution took, similar to what other engines offer (V8's d8 --times, etc.).

When the flag is passed, timing info is printed to stderr so it doesn't interfere with script output on stdout:
$ boa --time script.js Hello, World!
Execution: 1.56ms

For modules, parsing and execution are measured separately:
$ boa --time -m module.mjs 42
Parsing: 1.82ms Execution: 1.27ms Total: 3.09ms

Without the flag, there's zero overhead — no Instant::now() calls and no extra output.

Changes

All changes are in cli/src/main.rs:

• Added --time boolean flag to the Opt struct (long flag only, since -t is already used by --trace)
• Wrapped context.eval() in evaluate_expr and evaluate_file with conditional Instant measurements using args.time.then(Instant::now)
• For the module path, Module::parse() and load_link_evaluate() are measured separately
• In evaluate_expr, context.run_jobs() is called unconditionally after eval so async work is always drained regardless of whether timing is enabled

No new dependencies - uses std::time::Instant which was already imported (as Duration) in the file.

Closes #4709

[codecov]

Codecov Report

❌ Patch coverage is 0% with 85 lines in your changes missing coverage. Please review.
✅ Project coverage is 57.05%. Comparing base (6ddc2b4) to head (3edfe6d).
⚠️ Report is 704 commits behind head on main.

Files with missing lines	Patch %	Lines
cli/src/main.rs	0.00%	85 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4711      +/-   ##
==========================================
+ Coverage   47.24%   57.05%   +9.81%     
==========================================
  Files         476      552      +76     
  Lines       46892    60568   +13676     
==========================================
+ Hits        22154    34559   +12405     
- Misses      24738    26009    +1271     

☔ 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.

[github-actions]

Test262 conformance changes

Test result	main count	PR count	difference
Total	52,862	52,862	0
Passed	49,505	49,505	0
Ignored	2,261	2,261	0
Failed	1,096	1,096	0
Panics	0	0	0
Conformance	93.65%	93.65%	0.00%

[Nakshatra480]

Hey @jedel1043 just following up here 🙂

This adds a --time flag to the Boa CLI that prints parsing and execution timings to stderr (so it won’t mix with script stdout).

It measures parse vs. execution separately for modules and has no runtime overhead when the flag isn’t used.

Would appreciate a review from someone on the CLI/runtime side, happy to make any changes if needed.

[jedel1043]

Nice! I have a very small suggestion.

[jedel1043]

I kinda feel like this pattern of start = now, then executing something and elapsed = time can probably be simplified with a couple of structs:

struct Timer<'a> {
    name: &'static str,
    now: Instant,
    counters: &'a mut Vec<(&'static str, Duration)>
}

impl Drop for Timer<'a> {
    fn drop(&mut self) {
        self.counters.push((self.name, self.now.elapsed()));
    }
}


struct Counters {
    counters: Option<Vec<(&'static str, Duration)>>
}

impl Counters {
    fn new_timer(&mut self, name: &'static str) -> Option<Timer<'_>> {
        if let Some(counters) = &mut self.counters {
            Some(Timer {
              name,
              now: Instant::now(),
              counters
            })
        } else { None }
        
    }
}

impl Drop for Counters {
    // Code to print all timers
}

That way, to measure something you just do like

{
    let _timer = counters.new_timer("Parsing");
    // parse
}

// more code

[Nakshatra480]

Hey @jedel1043, thanks again for the review and the suggestions.

I’ve just pushed some updates to address your feedback:

• Timer & Counters struct: Implemented your suggestion! This makes the code super clean and ensures there's zero overhead when the flag isn't passed (we just use an Option and don't create Instants).
• Split timings everywhere: The timing is now consistently split into Parsing and Execution (or AST generation for --dump-ast) across all execution paths (module, script, and expression).
• Avoided print overhead: Moved print_timing() to happen completely independently from result/error printing, so we aren't measuring console I/O time anymore. I also made sure that async job execution (run_jobs()) is accurately included in the execution timer for consistency.
• Fixed -t flag conflict: Also noticed that the short -t flag was accidentally reassigned from --trace to --time in the previous iteration. I’ve reverted that so --trace keeps its short flag, avoiding any breaking changes there.

Let me know if this looks good to you or if there's anything else you'd like me to tweak!

[jedel1043]

Very nice work! Thank you!