added feature drop negative weights

[LauWien]

Please confirm that this PR has done the following:

• Tests Added
• Documentation Added
• Name of contributors Added to AUTHORS.rst
• Description in RELEASE_NOTES.md Added

Description of PR

This PR implements the proposed 8th solution of the issue #446

Write a warning when weights are negative and add a keyword-argument whether to drop (default) or not any values aggregated with negative weights.

By default any values with negative weights are getting now dropped. Additionally a warning's is beeing printed:

With the implemented boolean keyword argument drop_negative (default is True) this dropping can be overwritten by setting this keyword to False.

closes #446

[codecov]

Codecov Report

Merging #534 (1cdc0f3) into main (c9cb3ea) will increase coverage by 0.0%.
The diff coverage is 100.0%.

@@          Coverage Diff          @@
##            main    #534   +/-   ##
=====================================
  Coverage   93.5%   93.6%           
=====================================
  Files         48      48           
  Lines       5267    5285   +18     
=====================================
+ Hits        4929    4947   +18     
  Misses       338     338           

Impacted Files	Coverage Δ
pyam/core.py	92.6% <ø> (ø)
pyam/_aggregate.py	99.1% <100.0%> (+<0.1%)	⬆️
tests/test_feature_aggregate.py	98.8% <100.0%> (+<0.1%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c9cb3ea...1cdc0f3. Read the comment docs.

[danielhuppmann]

Thank you @LauWien, really nice draft! A few comments inline on more efficient code and better documentation, and a suggestion for a more elaborate test.

[danielhuppmann]

FYI, I also added the statement "closes #446" to the PR description - so when this PR is merged, the issue will be closed automatically

[danielhuppmann]

Thank you @LauWien!

I think that there is a flaw in your implementation, see the inline comment.

Also, the docstrings of pyam are not consistent, sometimes it's

arg : type, default <some value>
    bla ba

sometimes the (correct numpy-docs) style

arg : type, optional
    Bla ba

The idea is to pragmatically harmonize whenever a function is changed or updated - could you please do this for the docstring of the aggregate_region() function?

[danielhuppmann]

Thank you Laura - can you please rebase and implement the other to-dos from the description (and then mark them as closed): adding yourself to the list of authors of the package and adding one line to the release notes?

[danielhuppmann]

Thanks you, @LauWien, for your first contribution to the IAM community! Congrats!