Fixed the array types when interpolate_shells > 0. Now returns the p…

[Rodot-]

When interpolate_shells > 0, the formal integral broke with an attribute error claiming that numpy arrays had no attribute "values"
Issue #1463

I've corrected the return types of one of the functions so that they are no longer numpy arrays.

Description

When interpolate_shells > 0, interpolate_integrator_quantities would change the type of electron_densities_integ, tau_sobolev_integ, and Jredlu from pandas dataframes or series to numpy arrays due to the fact that scipy.interp1d always returns a numpy array. This was problematic because the formal_integral Cython wrapper access the .values attributes of these arrays assuming they were pandas objects. I've corrected this so now these arrays are set to have the correct type.

Motivation and Context

Now interpolate_shells can have values greater than 0 without failing.

How Has This Been Tested?

• Testing pipeline
• Reference Data Comparison following these instructions
• Other (please describe)

Screenshots (if appropriate):

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• None of the above (please describe)

Checklist:

• My code follows the code style of this project
• My change requires a change to the documentation
• I have updated the documentation accordingly
• I have built the documentation on my fork following these instructions
• I have assigned and requested two reviewers for this pull request

[codecov]

Codecov Report

Merging #1464 (a5f8151) into master (927c606) will decrease coverage by 1.13%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #1464      +/-   ##
==========================================
- Coverage   71.13%   70.00%   -1.14%     
==========================================
  Files          67       67              
  Lines        5523     5314     -209     
==========================================
- Hits         3929     3720     -209     
  Misses       1594     1594              

Impacted Files	Coverage Δ
...dis/tardis/montecarlo/montecarlo_numba/r_packet.py	23.15% <0.00%> (-7.83%)	⬇️
.../tardis/montecarlo/montecarlo_numba/interaction.py	24.32% <0.00%> (-5.68%)	⬇️
.../montecarlo/montecarlo_numba/single_packet_loop.py	24.56% <0.00%> (-3.78%)	⬇️
...rdis/tardis/montecarlo/montecarlo_numba/vpacket.py	15.00% <0.00%> (-2.48%)	⬇️
tardis/tardis/plasma/properties/util/macro_atom.py	30.00% <0.00%> (-2.26%)	⬇️
tardis/tardis/widgets/sdec_plot.py	14.35% <0.00%> (-1.94%)	⬇️
...rdis/montecarlo/montecarlo_numba/tests/conftest.py	91.11% <0.00%> (-1.88%)	⬇️
tardis/tardis/plasma/properties/nlte.py	39.21% <0.00%> (-1.74%)	⬇️
...s/tardis/montecarlo/montecarlo_numba/macro_atom.py	43.75% <0.00%> (-1.71%)	⬇️
tardis/tardis/montecarlo/spectrum.py	70.00% <0.00%> (-1.70%)	⬇️
... and 29 more

---

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 927c606...d0c371e. Read the comment docs.

[Rodot-]

Should also be of note: This bug really demonstrates the need for formal integral tests

[chvogl]

Thanks for fixing my blunder ;)

[chvogl]

I also agree that we should include a test with interpolate_shells > 0, which is the setting people should be using.

[chvogl]

I took the liberty to merge this without a second review. The changes are minimal and fix an urgent problem.