Remove unsafe usage

[CathalMullan]

Fuzzers work!

Need to think about how to optimize the UTF-8 checks.

Should we delay UTF-8 checks until post-inner-search? Or would that result in unnecessary work?

Could we remove the UTF-8 check after percent-decoding? Since we never actually return the provided decoded path.

Could we still use unsafe for the 'key'?

[codspeed-hq[bot]]

CodSpeed Performance Report

Merging #137 will degrade performances by 10.43%

_{Comparing 136-remove-unsage-usage-from-parameter-extraction (1b5d420) with main (96e9b7c)}

Summary

❌ 2 regressions ✅ 14 untouched benchmarks

:warning: Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Benchmark	main	136-remove-unsage-usage-from-parameter-extraction	Change
❌	matchit benchmarks/wayfind	11.8 µs	13.1 µs	-9.88%
❌	path-tree benchmarks/wayfind	58.9 µs	65.7 µs	-10.43%

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 73.26733% with 27 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/node/search.rs	56.89%	9 Missing and 16 partials :warning:
src/router.rs	75.00%	1 Missing and 1 partial :warning:

Files with missing lines	Coverage Δ
src/errors/path.rs	100.00% <ø> (ø)
src/errors/search.rs	100.00% <100.00%> (ø)
src/path.rs	100.00% <100.00%> (ø)
src/router.rs	90.83% <75.00%> (-0.70%)	:arrow_down:
src/node/search.rs	88.00% <56.89%> (-10.65%)	:arrow_down:

... and 1 file with indirect coverage changes