Py-pdf-parser

Added
- Published to PyPI as py-pdf-parser.
- Documentation is now hosted [here](https://py-pdf-parser.readthedocs.io/en/latest/). ([#71](https://github.com/jstockwin/py-pdf-parser/pull/71))
- Added new examples to the documentation. ([74](https://github.com/jstockwin/py-pdf-parser/pull/74))
- Font filtering now caches the elements by font. ([73](https://github.com/jstockwin/py-pdf-parser/pull/73)) (updated in [#78](https://github.com/jstockwin/py-pdf-parser/pull/78))
- Font filtering now caches the elements by font. ([73](https://github.com/jstockwin/py-pdf-parser/pull/73))
- The visualise tool now draws an outline around each section on the page. ([69](https://github.com/jstockwin/py-pdf-parser/pull/69)) (updated in [#80](https://github.com/jstockwin/py-pdf-parser/pull/80))


Changed
- This product is now complete enough for the needs of Optimor Ltd, however `jstockwin` is going to continue development as a personal project. The repository has been moved from `optimor/py-pdf-parser` to `jstockwin/py-pdf-parser`.

Added
- It is now possible to specify `font_size_precision` when instantiating a PDFDocument. This is the number of decimal places the font size will be rounded to. ([60](https://github.com/jstockwin/py-pdf-parser/pull/60))
- `extract_simple_table` now allows extracting tables with gaps, provided there is at least one full row and one full column. This is only the case if you pass `allow_gaps=True`, otherwise the original logic of raising an exception if there a gap remains. You can optionally pass a `reference_element` which must be in both a full row and a full column, this defaults to the first (top-left) element. ([57](https://github.com/jstockwin/py-pdf-parser/pull/57))

Changed
- Font sizes are now `float` not `int`. The `font_size_precision` in the additions defaults to 1, and as such all fonts will change to have a single decimal place. To keep the old behaviour, you can pass `font_size_precision=0` when instantiating your PDFDocument.

Fixed
- Improved performance of `extract_simple_table`, which is now much faster. ([65](https://github.com/jstockwin/py-pdf-parser/pull/65))

Added
- Initial version of the product. Note: The version is less than 1, so this product should not yet be considered stable. API changes and other breaking changes are possible, if not likely.