Reenable FLUX option and code for reaction flux computation & output

[msl3v]

Hey y'all @yantosca @RolfSander

I'm working with some folks who would like to have the fluxes reenabled in KPP. I'm going to start a separate branch and resurrect that code.

[msl3v]

I added branch '3.0.0-rc.4.flux'. It re-enables reaction flux. For reference, it mimics the way mecca does/used to do it: creating a bunch of dummy species for each reaction. But it does it internally rather than with a script.

[msl3v]

Issue: All of the flux species will impact the error norm computation because they are all initialized to zero. This buoys back up my attempt to fix this earlier. I didn't like how I did it before (as indicated in the associated comments). I think the best thing to do is simply

1. Force vector tolerances
2. and preset the ABSTOL() for the flux & family species to a high value. Maybe some factor above the concentration of the most abundant species.

The question is "where to put the presetting code?" I don't think _Parameters. is good because ABSTOL() is not a parameter.

I await your thoughts.

[msl3v]

I could add code to create species vector subsets, e.g. vec_RR would include all the flux species index values and could be used thus:
ATOL(vec_RR) =

Same can be done for families
ATOL(vec_FAM) =

These vectors could be set as parameters then.

[msl3v]

Turns out I'd already done vec_RR(), only it is called FLUX_MAP().
I added the equivalent of vec_FAM() and called it FAM_MAP().
So now you can use this vector to interact with all the Prod/Loss species

Added a new branch for it: 3.0.0-rc.4.flux.famMap