Fix problem with output data overflowing output array

[stscirij]

This is the same fix that was applied to STIS CCD data in #47. I didn't expect it to be a problem for COS data, but we have seen some NUV data overflow the output array and this small change fixes the problem

[stscirij]

This will result in one extra element at the long-wavelength end of each of the arrays, but all the other elements should be identical to those generated without this fix

[jotaylor]

@stscirij Curiously, I obtained identical results when I compared products from this branch with main. I used target AV-14 which has G185M data. Is that expected? I was expecting the data arrays to be one element longer.

[jotaylor]

Oops, disregard my previous comment. I don't think I properly checked out this branch when making the comparison products. Now I see the addition of another element in the data arrays. Otherwise the spectra appear identical.