Mindstorm

This version revises `mindstorm.statplot.plot_swarm_bar` to make it easier to customize properties of the points, bars, and error bars.

* The `plot_swarm_bar` function now takes `point_kws` and `bar_kws` arguments to set arguments for point and bar plots, respectively.
* The `capsize` input has been removed. Set this using `bar_kws={'capsize': mycapsize}` instead.

This version adds a `statplot` module for statistical visualization.

* Added `statplot.plot_swarm_bar` for visualizing the mean of one or more conditions along with 95% confidence intervals and the individual points in each condition using a swarm plot or strip plot. Generally supports the various options for Seaborn bar plots, such as plotting bars in different hues and creating groups of bars.

This version makes visualization of dissimilarity data more flexible.

* Changes plotting functions to include an optional `embedding` input. This can be any operation that supports the scikit-learn API, such as other dimensionality reduction techniques like tSNE.
* Renames some plotting functions to remove `mds`, as they now also support other dimensionality reduction techniques.
* Automatically adjusts the plot orientation for consistency across runs.

This version adds a new permutation test that was omitted from v0.5.0.

This version adds more options for working with dissimilarity matrices.

* Rank transformations when running partial representational similarity analysis are now optional.
* Plotting of similarity matrices is now supported; use `checks=False` when plotting similarity (instead of dissimilarity) using `mindstorm.dsmplot.plot_dsm`.

This version adds a module for creating plots of dissimilarity matrices and multidimensional scaling plots.