Vega Strike Code

Refactor Split CSV code a bit to also support split MPL

Fix defaults for functionality columns, make them "fully functional" to
allow omitting those columns in static datasets.

Some more split CSV fixes, fixes an issue when there are more than two CSV files listed.

Make ModUnitsCSV option be explicit (no longer defaulting to
modunits.csv), it allows us to bail with an error when we can't open a
referenced CSV, making debugging easier. Add some debugging output for
merging CSVs, and one tiny bugfix in column merging

Preliminar (not very tested) support for split CSV files.
De-hardcoded unit.csv, now there's data/ModUnitCSV and data/UnitCSV
config options. WARNING: ModUnitCSV was UnitCSV before.

only include parallelize loops when we want more than one

py3 hates multiple calls to the init functions for builtin python modules

some tab in .py issues and mostly killing dos endlines. because they need to die and i'm sick of them being ignored

Meta-vectorizer, version 1.
Applied to particle updates, as a POC.

I didn't see that it was a double tab ...