[newchem-cpp] calc_temp1d_cloudy,cool1d_cloudy: fix z interpolation bug and light cleanup

[mabruzzo]

This PR does 2 things:

1. the first commit fixes a redshift interpolation bug that I had accidentally introduced when I previously factored find z_index out of calc_temp1d_cloudy and cool1d_cloudy
   • I had been intending to write some explicit unit tests to check the control flow so that this sort of case never happens again, but never quite got to it. (And I think it's extremely important for us to merge in these bugfixes -- before we forget about them)
2. Subsequent commits lightly clean up calc_temp1d_cloudy. Importantly we clean up the function enough that we no longer require any heap allocations.
   • it also makes one or two similar tweaks to cool1d_cloudy and cool1d_cloudy_old_tables (but those are not the focus of this PR)

[brittonsmith]

@mabruzzo, I see you left a note about doing a few others things after PR #384 was merged. I'll let you take care of those and this merger conflict and then review.