fix: use same UNC path normalization logic with libuv

[sapphi-red]

fixes #295

[Boshen]

Requesting @Brooooooklyn because I don't know anything about windows.

[codspeed-hq]

CodSpeed Performance Report

Merging #306 will not alter performance

_{Comparing sapphi-red:fix/use-same-unc-path-normalization-logic-with-libuv (c8e2b09) with main (ebc6c36)}

Summary

✅ 3 untouched benchmarks

[Brooooooklyn]

@sapphi-red Please wait a moment while I verify if this custom canonicalize implementation is truly optimized on Windows

[sapphi-red]

@Brooooooklyn It seems cargo bench is slower on my machine.
Before (without sapphi-red#1)

resolver/single-thread  time:   [143.72 µs 148.36 µs 154.09 µs]
Found 8 outliers among 100 measurements (8.00%)
  2 (2.00%) high mild
  6 (6.00%) high severe
resolver/multi-thread   time:   [117.85 µs 119.70 µs 121.46 µs]
Found 7 outliers among 100 measurements (7.00%)
  4 (4.00%) high mild
  3 (3.00%) high severe
Benchmarking resolver/resolve from symlinks: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 301.3s, or reduce sample count to 10.
resolver/resolve from symlinks
                        time:   [2.9440 s 2.9456 s 2.9473 s]
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild

After (with sapphi-red#1)

resolver/single-thread  time:   [139.14 µs 144.00 µs 151.83 µs]
                        change: [−20.919% −3.8025% +16.565%] (p = 0.71 > 0.05)
                        No change in performance detected.
Found 12 outliers among 100 measurements (12.00%)
  4 (4.00%) high mild
  8 (8.00%) high severe
resolver/multi-thread   time:   [112.48 µs 114.09 µs 115.69 µs]
                        change: [−7.9616% −2.9621% +2.4978%] (p = 0.29 > 0.05)
                        No change in performance detected.
Found 5 outliers among 100 measurements (5.00%)
  1 (1.00%) high mild
  4 (4.00%) high severe
Benchmarking resolver/resolve from symlinks: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 743.3s, or reduce sample count to 10.
resolver/resolve from symlinks
                        time:   [7.4256 s 7.4356 s 7.4479 s]
                        change: [+152.08% +152.43% +152.85%] (p = 0.00 < 0.05)
                        Performance has regressed.
Found 5 outliers among 100 measurements (5.00%)
  1 (1.00%) high mild
  4 (4.00%) high severe

[Brooooooklyn]

it seems cargo bench is slower on my machine.

Me too, It seems that we can simply use node_compatible_raw_canonicalize(fs::canonicalize(&path)) under Windows.

[sapphi-red]

I reverted some code to use fs::canonicalize for Windows.

[Boshen]

Please let me know when ready to merge.

[sapphi-red]

It's ready to merge 👍