Define ParseError as a dataclass

[pgiraud]

Fixes #152

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.60%. Comparing base (5c45afe) to head (6da3a65).
⚠️ Report is 7 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #153      +/-   ##
==========================================
- Coverage   97.61%   97.60%   -0.01%     
==========================================
  Files          10       10              
  Lines        1214     1213       -1     
==========================================
- Hits         1185     1184       -1     
  Misses         29       29              

☔ View full report in Codecov by Harness.
📢 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.

[dlax]

I'm not sure defining exception classes as dataclass is a good idea; I've never seen this pattern. In particular, I am worried about the fact that the super-class constructor is no longer called, so that the args attribute, part of BaseException API, is empty. This might not be a very big deal in practice depending on what users do with the exception value, but this does not seem very clean.