Fixed all unit tests.
Removed constant padding added around tables in the last step of the
initial discovery mode of the stream algorithm.
Replaced it with a heuristic that attempts to expand the table up while
respecting columns identified so far.
Updated unit tests to reflect new behavior, improved rejection of
extraneous information in few cases.
Added unit test covering a use case where the header has vertical test.
Made improvements to better support vertical text in tables.
* [MRG] Create a new figure and test each plot type #127
- move `plot()` to `plotting.py` as `plot_pdf()`
- modify plotting functions to return matplotlib figures
- add `test_plotting.py` and baseline images
- import `plot_pdf()` in `__init__`
- update `cli.py` to use `plot_pdf()`
- update advanced usage docs to reflect changes
* Change matplotlib backend for image comparison tests
* Update plotting and tests
- use matplotlib rectangle instead of `cv2.rectangle` in
`plot_contour()`
- set matplotlib backend in `tests/__init__`
- update contour plot baseline image
- update `test_plotting` with more checks
* Update plot tests and config
- remove unnecessary asserts
- update setup.cfg and makefile with `--mpl`
* Add to
* Add tolerance
* remove text from baseline plots
update plot tests with `remove_text`
* Change method name, update docs and add pep8
* Update docs