Including atmosphere, qe, reflectivity in VISTA filters

[stevet40]

Fixes #40 (but at what cost?)

This is a much bigger change than #39, so it would probably be good for someone to independently verify what I've done here.

I have downloaded the data from the ESO webpage, and used it to construct combined transmission curves for VISTA/VIRCAM. These include the Paranal model atmosphere (PWV=1mm, airmass=1), the quantum efficiency, and the mirror reflectivity for M1 and M2. The combined curves I'm proposing are shown below, compared to what we have previously (which was filter only).

My proposed change here is to promote these new curves to the names vista_vircam_*.par, and to retain what we had previously as something like vista_vircam_*_FILTER.par. I've tried to improve the comments in the various files, so it's hopefully unambiguous what I did in each case.

Two things where I made a decision (that others may disagree with):

1. I used the PWV=1mm, airmass=1 atmosphere (I think they used something like this in SVO, and I think the EAZY vista+atm filters used this, but I haven't verified/compared in detail). There is a second atmosphere model with PWV=3mm, airmass=2.
2. I truncated the curves to the range of the tabulated QE (assuming the QE basically drops to 0 outside this range, which may not be entirely true).

Let me know what you think!

[bd-j]

I think this looks good, and the comments seem clear about what was done. The QE is probably not totally zero outside the range, but it does look to be dropping rapidly enough that this is a reasonable choice to make. I can update the filters/README.md file when this is merged.

[bd-j]

Thanks @stevet40 !!!