refactor(oxfmt): simplify wrap_format_embeded

[Dunqing]

call_async does the same thing as call_with_return_value + channel, so use call_async to simplify it and reduce some unnecessary overhead

Benchmark

I benchmarked this PR locally, and the results indicate that it is 9% faster than before in the Outline repo.

EDIT: Maybe just a noise since Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interference from other programs. was displayed.

Setup complete! Run 'pnpm run bench' to start benchmarking.
=========================================
JavaScript/TypeScript Formatter Benchmark
=========================================

Formatters: Prettier, Biome, Oxfmt

Starting benchmark with:
- 3 warmup runs
- 10 benchmark runs
- Git reset before each run


=========================================
Benchmarking parser.ts (single large file)
=========================================
Benchmark 1: biome
  Time (mean ± σ):      69.5 ms ±   0.9 ms    [User: 55.4 ms, System: 9.7 ms]
  Range (min … max):    68.3 ms …  71.2 ms    10 runs
 
Benchmark 2: oxfmt-new
  Time (mean ± σ):      30.7 ms ±   0.3 ms    [User: 23.7 ms, System: 11.0 ms]
  Range (min … max):    30.3 ms …  31.3 ms    10 runs
 
Benchmark 3: oxfmt-old
  Time (mean ± σ):      30.4 ms ±   0.3 ms    [User: 23.6 ms, System: 10.6 ms]
  Range (min … max):    30.1 ms …  30.9 ms    10 runs
 
Summary
  oxfmt-old ran
    1.01 ± 0.01 times faster than oxfmt-new
    2.29 ± 0.04 times faster than biome

=========================================
Benchmarking Outline repository
=========================================
Benchmark 1: biome
  Time (mean ± σ):     154.6 ms ±   6.0 ms    [User: 825.2 ms, System: 555.8 ms]
  Range (min … max):   147.5 ms … 164.0 ms    10 runs
 
Benchmark 2: oxfmt-new
  Time (mean ± σ):      65.8 ms ±   2.3 ms    [User: 196.0 ms, System: 265.3 ms]
  Range (min … max):    60.7 ms …  68.8 ms    10 runs
 
Benchmark 3: oxfmt-old
  Time (mean ± σ):      72.0 ms ±  17.7 ms    [User: 195.0 ms, System: 277.3 ms]
  Range (min … max):    62.6 ms … 121.1 ms    10 runs
 
  Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs.
 
Summary
  oxfmt-new ran
    1.09 ± 0.27 times faster than oxfmt-old
    2.35 ± 0.12 times faster than biome

[Dunqing]

• refactor(oxfmt): simplify wrap_format_embeded #15798 👈 (View in Graphite)
• perf(formatter): use ArenaVec and ArenaBox #15420
• feat(allocator/vec): add Vec::into_bump_slice method #15770
• main

---

How to use the Graphite Merge Queue

Add either label to this PR to merge it via the merge queue:

• 0-merge - adds this PR to the back of the merge queue
• hotfix - for urgent hot fixes, skip the queue and merge this PR next

You must have a Graphite account in order to use the merge queue. Sign up using this link.

An organization admin has enabled the Graphite Merge Queue in this repository.

Please do not merge from GitHub as this will restart CI on PRs being processed by the merge queue.

This stack of pull requests is managed by Graphite. Learn more about stacking.

[leaysgur]

Just FYI, you may or may not know:

• The outline repository contains 26 files that include css`...`
• But there are only 4 of them do not contain substitution
  • = cases where the current oxc_formatter calls Prettier

So while the benchmark might not be very reliable, I still think this approach is better. 😅

[leaysgur]

Merge activity

• Nov 18, 6:28 AM UTC: The merge label '0-merge' was detected. This PR will be added to the Graphite merge queue once it meets the requirements.
• Nov 18, 6:28 AM UTC: leaysgur added this pull request to the Graphite merge queue.
• Nov 18, 6:35 AM UTC: Merged by the Graphite merge queue.