fix(parser): Switch literal impl's from `one_of` to `literal`

[epage]

Originally, this was going to be done for performance reasons but when comparing this to the previous commit, tag(char) / char slowed down dramatically in the next_slice benchmark.

However, json benchmark was unchanged. This also (slightly) simplifies traces, is more direct in what it says, and does still speed things up for UTF-8-exclusive chars, so going forward with it.

Fixes #418 Fixes #427

BREAKING CHANGE: char parers now always return char as the output and now require I: Compare<char>

BREAKING CHANGE: u8 parers now require I: Compare<u8>

[coveralls]

Pull Request Test Coverage Report for Build 7818956700

• -4 of 5 (20.0%) changed or added relevant lines in 2 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.1%) to 42.7%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/ascii/mod.rs	0	2	0.0%
src/parser.rs	1	3	33.33%
<!--	Total:	1	5	20.0%	-->

Totals
Change from base Build 7818240433:	0.1%
Covered Lines:	1319
Relevant Lines:	3089

---

💛 - Coveralls

[codecov[bot]]

Codecov Report

Attention: 4 lines in your changes are missing coverage. Please review.

Comparison is base (3373ba1) 42.60% compared to head (502a36f) 42.57%.

Files	Patch %	Lines
src/ascii/mod.rs	0.00%	2 Missing :warning:
src/parser.rs	33.33%	2 Missing :warning:

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #464 +/- ## ========================================== - Coverage 42.60% 42.57% -0.04% ========================================== Files 18 18 Lines 3089 3089 ========================================== - Hits 1316 1315 -1 - Misses 1773 1774 +1 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.