fix(cvfdutil): polygon area and centroid calculations now use shapely

[langevin-usgs]

• Changed area_of_polygon and centroid_of_polygon to use shapely calculations
• Close bug: possible error detected in cvfd_utils cell centroids function shapefile_to_xcyc #2150
• Added test, which uses a polygon that was failing with the previous implementation

Decided to use shapely calculations instead of our own flopy calculations. This should much more robust for real-world models, such as the one described in #2150.

[codecov]

Codecov Report

Attention: Patch coverage is 62.50000% with 3 lines in your changes are missing coverage. Please review.

Project coverage is 71.0%. Comparing base (029a4e1) to head (a9906c1).
Report is 2 commits behind head on develop.

❗ Current head a9906c1 differs from pull request most recent head c942148. Consider uploading reports for the commit c942148 to get more accurate results

Additional details and impacted files

@@            Coverage Diff            @@
##           develop   #2165     +/-   ##
=========================================
+ Coverage     68.3%   71.0%   +2.7%     
=========================================
  Files          260     260             
  Lines        57947   58128    +181     
=========================================
+ Hits         39611   41318   +1707     
+ Misses       18336   16810   -1526     

Files	Coverage Δ
flopy/utils/cvfdutil.py	80.7% <62.5%> (-2.6%)	⬇️

... and 178 files with indirect coverage changes

[wpbonelli]

Looks good. I wonder if there are other geometry calculations we could consider delegating to shapely. Maybe worth a scan at some point