Fix incorrect end position in read_to_end and read_text

[Mingun]

When working on #766 I found this bug. When trim_text_start is set to true, the trailing spaces could not be included in text returned by the read_text method and not counted in the end field of the returned span from read_to_end family of methods. If you read text content of such XML:

<tag>  <nested>  </tag>

then the trailing space would be dropped. This is because we detect end position of the span before we read </tag>. When start spaces are trimmed we have 3 events: Start, Empty, End, and buffer_position() after reading each of them are:

<tag>  <nested>  </tag>
     ^         ^ |     ^ - buffer_position() after each event
     ^           ^       - expected Span of the read_to_end() methods

Spaces just skipped and do not update buffer_position() which results to the incorrect end of the Span. This is happened only when the last meaningful event before the closing End event is not a Text event.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 60.37%. Comparing base (7558577) to head (930d1b2).
Report is 54 commits behind head on master.

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #773      +/-   ##
==========================================
- Coverage   61.81%   60.37%   -1.45%     
==========================================
  Files          41       41              
  Lines       16798    16229     -569     
==========================================
- Hits        10384     9798     -586     
- Misses       6414     6431      +17     

Flag	Coverage Δ
unittests	60.37% <ø> (-1.45%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.