fix RuntimeWarning for nan values

[cshanahan1]

This addresses #366

[jdavies-st]

Instead of changing the data, which could have adverse side-effects, I think it might be better to allow the data to pass through as nans and then capture the warning, especially if it is expected as in this case. So something like:

import warnings

    with warnings.catch_warnings():
        warnings.filterwarnings(action="ignore",
                                message="invalid value encountered in remainder",
                                category=RuntimeWarning)
        lon = np.mod(lon, 360.0 * u.deg if nquant else 360.0)

Which uses a context manager to apply a very specific warnings ignore filter just within the with block.

[codecov]

Codecov Report

Merging #367 (cb7932e) into master (8b8a2bd) will increase coverage by 0.01%.
The diff coverage is 100.00%.

❗ Current head cb7932e differs from pull request most recent head 7991e3b. Consider uploading reports for the commit 7991e3b to get more accurate results

@@            Coverage Diff             @@
##           master     #367      +/-   ##
==========================================
+ Coverage   86.34%   86.35%   +0.01%     
==========================================
  Files          21       21              
  Lines        3624     3627       +3     
==========================================
+ Hits         3129     3132       +3     
  Misses        495      495              

Impacted Files	Coverage Δ
gwcs/geometry.py	98.01% <100.00%> (+0.06%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 8b8a2bd...7991e3b. Read the comment docs.

[jdavies-st]

Looks good. Just needs a change log entry.

[nden]

Needs a change log

[nden]

I think this is more readable if instead of catching the warning the call is modified as in

lon = np.mod(lon, 360.0 * u.deg if nquant else 360.0, where=np.isfinite(lon), out=lon)