Polyagamma

- The API has been changed from a class-based to a function-based one. See README file for more info,
- A new "Saddle Point Approximation" sampling method has been added. It can be accessed using `method="saddle"`.
- The hybrid sampler has been updated.
- A normal approximation sampler for when `h` is large is now available. This functionality is not directly available in the public API and only usable via the hybrid sampler.

- The hybrid sampler has been updated.
- A normal approximation sampler for when `h` is large is now available. This functionality is not directly available in the public API and only usable via the hybrid sampler.
- Speed of the `saddle` method has been improved with use of a different solver.

- The API has been changed from a class-based to a function-based one. See `README` file for more info,
- A new "Saddle Point Approximation" sampling method has been added. It can be accessed using `method="saddle"`.

This pre-release is available on PyPI and can be installed with `pip install polyagamma==1.0.0b1`

First "stable" release. See README for Details.

- The alternate sampling method was added.
- A new keyword argument `method` is available to select a sampling algorithm. Availble sampling methods are `devroye`, `gamma` and `alternate`.