Show labels of grouped pandas dataframe in one legend

[zxdawn]

Description

Proplot plots the legend for each label instead of listing labels on one legend.
As a result, several legends are stacked together.

Steps to reproduce

import proplot as plot

df = pd.DataFrame({'A' : ['foo', 'bar', 'foo', 'bar',
                          'foo', 'bar', 'foo', 'foo'],
                   'B' : ['one', 'one', 'two', 'three',
                       'two', 'two', 'one', 'three'],
                   'C' : np.random.randn(8),
                   'D' : np.random.randn(8)})

fig, axs = plot.subplots()
df.groupby('A')['C'].plot(legend=True, ax=axs)

Expected behavior:

Actual behavior:

Equivalent steps in matplotlib

import matplotlib.pyplot as plt

df = pd.DataFrame({'A' : ['foo', 'bar', 'foo', 'bar',
                          'foo', 'bar', 'foo', 'foo'],
                   'B' : ['one', 'one', 'two', 'three',
                       'two', 'two', 'one', 'three'],
                   'C' : np.random.randn(8),
                   'D' : np.random.randn(8)})

df.groupby('A')['C'].plot(legend=True)
plt.show()

Proplot version

0.6.3

[lukelbd]

This probably has something to do with the fact that proplot's Axes.legend() permits drawing multiple legends in a subplot with multiple legend() calls, while matplotlib's Axes.legend() does not. Thanks for the report, using proplot axes with pandas/xarray/seaborn plotting functions is something I hope to more rigorously support in the future.

[lukelbd]

Fixed this in #254. Also inspired a refactoring/cleanup of proplot's legend internals