inspector: fix StringUtil::CharacterCount for unicodes

[legendecas]

StringUtil::CharacterCount should return the length of underlying
representation storage of a protocol string.

StringUtil::CharacterCount is only used in DictionaryValue
serialization. Only Network.Headers is an object type, represented
with DictionaryValue.

[nodejs-github-bot]

Review requested:

• @nodejs/gyp
• @nodejs/inspector
• @nodejs/cpp-reviewers

[nodejs-github-bot]

CI: https://ci.nodejs.org/job/node-test-pull-request/64768/

[lemire]

This goes back to at least this commit by @targos

5aaa7fe

(6 years ago)

where the code was...

size_t CharacterCount(const String& s) {
  icu::UnicodeString utf16 =
      icu::UnicodeString::fromUTF8(icu::StringPiece(s.data(), s.length()));
  return utf16.countChar32();
}

[lemire]

So it was always wrong?

[legendecas]

@lemire this function was never called. Only the new protocol Network.Headers uses this function, like mentioned in the OP.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 89.22%. Comparing base (04f147b) to head (2908bef).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main   #56788   +/-   ##
=======================================
  Coverage   89.22%   89.22%           
=======================================
  Files         663      663           
  Lines      191974   191974           
  Branches    36922    36919    -3     
=======================================
+ Hits       171282   171293   +11     
+ Misses      13561    13555    -6     
+ Partials     7131     7126    -5     

Files with missing lines	Coverage Δ
src/inspector/node_string.cc	60.00% <100.00%> (ø)
src/inspector/node_string.h	57.14% <ø> (ø)

... and 41 files with indirect coverage changes