Use bottleneck to calculate nanmean in statistics function

- pandas docs mention [this](https://pandas.pydata.org/docs/getting_started/install.html#recommended-dependencies):
> bottleneck: for accelerating certain types of nan evaluations. bottleneck uses specialized cython routines to achieve large speedups. If installed, must be Version 1.2.1 or higher.

- We can use bottleneck to calculate the nanmean, and speed up our describe function
- We should make this an **optional dependency**
- It seems that we might just need to install `bottleneck` and we should be able to get the speed improvement, but we should verify this.