add abstrct dataset

[ArneBinder]

In this PR, we add AbsTRCT - an argumentation-mining dataset, according to the description in #10.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (856ca5d) 94.72% compared to head (172fc14) 95.15%.
Report is 12 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #68      +/-   ##
==========================================
+ Coverage   94.72%   95.15%   +0.43%     
==========================================
  Files          19       21       +2     
  Lines        1289     1342      +53     
==========================================
+ Hits         1221     1277      +56     
+ Misses         68       65       -3     

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

[ArneBinder]

Looks very good, just some minor points, see below.

[idalr]

While the AbstRCT dataset has no fragmented spans, we directly converted the dataset to BratDocumentWithMergedSpans, as a single-variant version ("default") in 612466b.

Updates:

• 172fc14 fixed per comment below

[ArneBinder]

Looks very good, just some minors.