fixes for more edge cases in raw tokenizer

[mdelah]

This PR closes out the remaining issues from #35.

As discussed:

• Left and right angle brackets within <!-- and not part of the closing --> are considered part of the comment, and
• Right angle brackets are considered valid in attribute values (and no longer create a split in RawToken)

I also realized there was a similar issue with <? ... ?> tags, so made some fixes there too.

I had to rearrange RawToken a bit to get this to work efficiently. It now calls out to a new function findTokenEnd to locate the closing >, which returns -1 if it's not within the current buffer. The findTokenEnd has logic to skip past > that occur within comments or quoted values. It calls itself to skip past nested tags (such as may appear inside <!DOCTYPE [ ... ] >.

I took the opportunity to replace some of the byte-by-byte iteration with bytes.IndexByte, which gives a moderate performance boost in some of the benchmarks.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.54%. Comparing base (5d45a12) to head (e9c4ca0).

Additional details and impacted files

@@            Coverage Diff             @@
##           master      #43      +/-   ##
==========================================
- Coverage   99.00%   98.54%   -0.47%     
==========================================
  Files           2        2              
  Lines         302      343      +41     
==========================================
+ Hits          299      338      +39     
- Misses          2        4       +2     
  Partials        1        1              

☔ 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.