Allow reversing cumulative density when using kdeplot in distplot

Rationale: I can invert the cumulative histogram in matplotlib, so it would be nice if I could reverse also the result of kdeplot when using distplot.

[erinboyle]

I think this has real use @mwaskom . I’m trying to show what fraction of clients at my company have accumulated at least N datapoints. Cumulative in the normal direction is showing me how many clients have accumulated at most N datapoints – much less relevant to the question we’re interested in. I’m using plt.hist instead since they offer reverse, but far prefer seaborn’s formatting.

[lbeltrame]

That’s fine, I found a clean solution using statsmodels, so it should be good enough for me.