Summary

Hey, not sure if there is in fact an internal reason for this choice, but personally I find it odd
that plt.hexbin() uses a upper-case C as kwarg to assign custom values for aggregation in each bin.

In contrast, plt.scatter uses a lower-case c to assign values to each datapoint.

Proposed fix

I believe it would be good in the long run to switch from upper-case C to lower-case c in plt.hexbin()