Use Flag for improving the management of CodeCov

[lionelkusch]

Based on the flags documentation of CodeCov.
I try to improve the management of the coverage.

I will try next week to test it and see if it's working or not.

This is linked to the issue of the coverage of some Pull Requests are not passing because the dataset tests are not running.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.48%. Comparing base (7880eb1) to head (c03823d).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #343      +/-   ##
==========================================
+ Coverage   96.41%   96.48%   +0.06%     
==========================================
  Files          63       50      -13     
  Lines        6919     5943     -976     
==========================================
- Hits         6671     5734     -937     
+ Misses        248      209      -39     

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[lionelkusch]

I created Proof of concept: here
I mostly fix everything. The only missing point is that the dataset flag considers all the files. I don't understand why.

@rflamary
Looking more into details of the tests in the dataset folder, I remarked that these tests don't test all the functions in the folder and some tests outside of this folder test some specific functions of it.

If I limit the tests in the dataset folder, I will reduce the coverage of the folder by 1.7 % of the dataset folder.
The other solution is to run all the tests for keeping the same coverage but I don't really like this solution.

What do you think?

[lionelkusch]

I fixed the missing point.

After your answer, this PR can be merged.

[tgnassou]

LGTM