Gnssrefl

pushed Tim Dittmann's changes to make_meta.
Still some issues that need to be fixed.

Changed input on refl_zones for ellipsoidal height to be consistent with gnssir_input
Commandl line variable shoud be height, not el_height.

Fixed unset variable in the data windowing function so it always has a value (iAzim).

I updated gnssir and quickLook so that it now reports the azimuth of the lowest
elevation angle in the arc. Previously it used the average azimuth over the
entire rising or setting arc.

Changed gnss back to its original setting, which is getting the multi-GNSS final
GFZ orbits from CDDIS. I had sent those queries to rapid, not remembering that rapid
GFZ does not have beidou.

gnss-gfz directly downloads multi-GNSS orbits from GFZ. this is equivalent to gnss3 option.

for all four constellations, use gnss-gfz or gnss options.

Added script to download tide gauge data from Queensland ... download_qld.py

quickplt allows xaxis limits in MJD (i.e. it will use datetimes to make a nicer x-axis)

Fixed bug in query_modern in gps.py that did not check to see if Files subdirectory existed before
downloading station_pos.db.

I do not know WHY the docker does not provide the necessary file, as it should.

I incorrectly implemented the refraction correction when I wrote the newarcs
option of gnssir. This has been fixed with this version. I will be removing all
pypi versions that have the bug in it. Effectively, the refraction correction
was not being implemented at all. this will cause a bias between series computed before
and after. It has no impact on the SNR data files. Only on gnssir output (and the programs
that use those outputs, like subdaily, snowdepth, phase, and vwc).


changed WSV water level to 30 day downloads instead of 15

gnssir_input incorrectly defined e1 and e2 as integers ... fixed.

quickLook also incorrectly defined e1 and e2 as integers ... fixed

added more columns to subdaily outlier file...