Fixing error/warnings when writing to HDF5, circleRadius bug.

[astronomerritt]

Fixes #722.
Fixes #727

The NaturalNameWarning is now suppressed when writing to HDF5 when your object IDs start with a number (or don't otherwise fit PyTables' rather strict naming pattern ^[a-zA-Z_][a-zA-Z0-9_]*$). Unless the user is going to be using PyTables to look at the data, this won't be an issue - we can warn them about this in the docs.

While recreating this error, I also found a bug. At some point in the code -- I have no idea why or where, grepping didn't help -- the ObjID column is being changed to a Pandas dtype that PyTables doesn't support ('StringDtype'), leading to errors when writing out to HDF5.

This wasn't caught in the unit test because the unit test for PPOutWriteHDF5() just loads in some data from a CSV file then writes it out. Short of running all of Sorcha in test_PPOutWriteHDF5(), I'm not sure this COULD have been caught - might be an argument for having end-to-end tests for all the output formats.

I have inserted a line in PPOutWriteHDF5() that changes the type of that column to a standard 'str' instead.

ADDITIONAL FIX:
I have also fixed that one horrible lonely instance of circleRadius being used instead of circle_radius in PPPrintConfigsToLog(). Finally.

Review Checklist for Source Code Changes

• Does pip install still work?
• Have you written a unit test for any new functions?
• Do all the units tests run successfully?
• Does Sorcha run successfully on a test set of input files/databases?
• Have you used black on the files you have updated to confirm python programming style guide enforcement?

[codecov]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Files	Coverage Δ
src/sorcha/modules/PPOutput.py	84.78% <100.00%> (+1.44%)	⬆️
src/sorcha/modules/PPConfigParser.py	66.95% <0.00%> (ø)

📢 Thoughts on this report? Let us know!