Fix bug 407

[GabrielKS]

See bug #407 for a detailed description of the issue. This implements the proposed fix from that bug report, i.e., adds a special case for AbstractArray. I couldn't resist a bit more cleanup of now-redundant code as well.

[jd-lara]

makes sense. Now array objects have ref and size as fields.

[codecov]

Codecov Report

Attention: Patch coverage is 85.71429% with 1 line in your changes missing coverage. Please review.

Project coverage is 77.65%. Comparing base (627c8dd) to head (83da575).
Report is 3 commits behind head on main.

Files with missing lines	Patch %	Lines
src/utils/utils.jl	85.71%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #408      +/-   ##
==========================================
+ Coverage   77.63%   77.65%   +0.02%     
==========================================
  Files          69       69              
  Lines        5481     5474       -7     
==========================================
- Hits         4255     4251       -4     
+ Misses       1226     1223       -3     

Flag	Coverage Δ
unittests	77.65% <85.71%> (+0.02%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
src/utils/utils.jl	62.45% <85.71%> (+0.12%)	⬆️

[jd-lara]

@GabrielKS is it expected that this breaks PSY?

[GabrielKS]

@GabrielKS is it expected that this breaks PSY?

No, that was not expected (though strictly speaking it is only breaking the PSY tests…). I'll get into it.

[GabrielKS]

Okay, yesterday evening's implementation was fundamentally wrongheaded. We do need a special case for AbstractArray to avoid testing the fields, but we don't want it to go directly to the match function, we want it to iterate over the items and recurse — like we already had for Vector. Here I've adapted that existing Vector method so it works with arrays of arbitrary dimensionality. Should not break PowerSystems now.