Readgssi

- changed file naming convention (outfile parameter is no longer clobbered by the naming function)
- changed main python function to accept user-specified velocity threshold for pause correction function call

- added support for the "42000S" antenna name ([34](https://github.com/iannesbitt/readgssi/issues/34) from [GGDRriedel](https://github.com/GGDRriedel))
- added support for the "1.5/1.6GHz" antenna name ([39](https://github.com/iannesbitt/readgssi/issues/39))
- corrected a misuse of a great circle distance calculation using geopy ([38](https://github.com/iannesbitt/readgssi/issues/38) from [felippehsk](https://github.com/felippehsk))
- allow for graceful handling of empty GGA strings ([38](https://github.com/iannesbitt/readgssi/issues/38) from [felippehsk](https://github.com/felippehsk))
- added support for the "4105NR" antenna name ([21](https://github.com/iannesbitt/readgssi/issues/21) from [eyoabz](https://github.com/eyoabz))
- added limited support for the "300/800D" antenna name ([33](https://github.com/iannesbitt/readgssi/issues/33) from [alecomunian](https://github.com/alecomunian)). GSSI bug prevents this combination of antennas from full support, as the second antenna is named `'none'` in this case, but all readgssi functionality should still work
- fixed a previously unknown bug that did not separate multi-channel outputs into separate files (addresses [33](https://github.com/iannesbitt/readgssi/issues/33) from [alecomunian](https://github.com/alecomunian) and [#37](https://github.com/iannesbitt/readgssi/issues/37) from [h2maa](https://github.com/h2maa))
- changed license from AGPLv3 to GPLv3 as standard GPLv3 fits the purpose better

- fixed a problem that led to errors creating a conda skeleton
- explicitly specify base _e_ (`numpy.e`) in `SymLogNorm` call in `readgssi:plot.radargram`
- added the ability to output to any supported matplotlib format
- fixed `docs/make_plots.sh` and branched a `docs/make_readme_plots.sh` for README figure creation
- corrected docs text and `docs/_static` images that incorrectly applied FIR and BGR (or didn't apply them at all)
- added an example plot demonstrating the absolute value of gradient (`-A -c gray_r`) method

- fixed the read of the byte that contains `rh_system` and `rh_version` in the header ([26](https://github.com/iannesbitt/readgssi/issues/26))
- fixed sampling depth interpretation ([28](https://github.com/iannesbitt/readgssi/issues/28))
- added DZT file output ([29](https://github.com/iannesbitt/readgssi/pull/29))
- added documentation for displaying absolute value of down-trace gradient in plots
- fixed a bug that incorrectly displayed long axis units on stacked arrays
- added [conda installation method](https://readgssi.readthedocs.io/en/latest/installing.html#installing-from-anaconda-cloud) and streamlined installation documentation
- switched version handling to file (`_version.py`)
- fixed a bug that did not handle `rhf_depth == 0` cases correctly
- added support for the 3102 antenna (500 MHz)
- fixed a bug that incorrectly plotted TWTT
- updated examples in Readme and fixed the autogen script (`make_plots.sh`)
- corrected an error with how `setup.py` got the version number, which prevented the creation of a conda skeleton

- merged [19](https://github.com/iannesbitt/readgssi/pull/19) from [teshaw](https://github.com/teshaw) which allows the user to specify partial read parameters for DZT files
- reintroduced a testing feature which creates CSV files from DZG records, in order to easily import location information to GIS
- added support for the StructureScan Mini XT which suffered from the unfortunate fate of having an antenna named 'none'
- improved handling of multiple user-specified frequencies
- adding the ability to plot the absolute value of the down-trace gradient, which can increase faint layer visibility at the cost of wave front resolution
- now use `rf_top` to correct depths, a major bug in previous versions
- limited distance normalization console output to one message every ten chunks (in the future, this will be controlled with a \r statement)
- added pause correction function to fix DZG/DZT trace number offset when collection is paused and GPS is active
- added ability to plot user marks as vertical lines (similar to the output on the controller itself)
- corrected module call documentation for plotting ([23](https://github.com/iannesbitt/readgssi/issues/23))

- fixed [14](https://github.com/iannesbitt/readgssi/issues/14) which incorrectly asserted that user-set samples per meter was negative when it was not (sign was flipped) but processed the file correctly regardless
- moved zoom calculation and automatic file naming to their own functions
- added sphinx autodoc descriptions to some functions, in order to work towards solving [9](https://github.com/iannesbitt/readgssi/issues/9)
- added comments to datetime parsing function
- fixed sampling frequency calculations to the value of `epsr` set in the header, which will prevent perceived frequency wander in frequency-based filters if the user sets `epsr`
- changed the behavior of the readdzt script to automatically search for a DZG file of the same name, and load that data if possible. this should allow python users to read all three variables (header, arrays, gps) if GPS exists.
- fixed command line control of DPI
- switched to "gray" colormap from "Greys". "Greys" rendered opposite polarity [18](https://github.com/iannesbitt/readgssi/issues/18)