Wfc3tools

------------------
- Major updates to the readthedocs documentation, particularly for UVIS pipeline and new full-well saturation [68, 83]
- Updated some Python APIs for ``calwf3`` component wrappers and explained parameters [69, 71, 85]
- Updated embedsub.py to ensure all input extensions are included in the output file [70]
- Reversed the use of column and row parameters when accessing image data and updated example for pstack [73]
- Modified how an image section is specified and reversed the use of column and row parameters for pstat [75]
- Dropped support for Python 3.8 [79]

------------------
- fixed matplotlib plotting issue, added explicit draw and interaction plots
- updates to package setup, removed support for TEAL interface [63]
- changes to documentation infrastructure and text [64]
- fixed bug in sub2full.py that required input file to be in pwd, now a file path can be supplied [64]
- removed 'version' argument (which corresponds to the command line -r flag) from calwf3 python interface [64]
- updates to module docstrings [64]

----------------
- Documentation-only updates corresponding to calwf3 3.4.1 release
- travis sphinx install update for testing

----------------
- bugfix for input string to calwf3 executables

----------------
This release incorporates documentations changes for CALWF3 which will increase from v3.3 to v3.4 with the installation of HSTDP 2016.2 in the Space Telescope Data processing system. Included are:

- updated information on CTE correction for subarray images
- updated information on Sink pixel detection for subarray images
- updated interaction with the Jupyter notebook when interacting with the calwf3 executables
- improved logging and error return information from the kernel

----------------
- package reorganization
- addition of output capture/logging to the calwf3 executables, help display text in jupyter notebook
- moved the documentation to readthedocs and removed the local build, users will now be pointed to RTD from the display_help() function