Visualization breaks when intent is not part of resulting table

[dorisjlee]

When displaying the visualization, it is possible that the intent is not present in the resulting data table. For example in the case of the index, the intent is propagated over to the resulting table, but the attribute Attrition is no longer in the table. When the intent is not present in the dataframe, we should handle the error by falling back to the default visualizations.

df = pd.read_csv("https://raw.githubusercontent.com/lux-org/lux-datasets/master/data/employee.csv")
df.intent = ["Attrition"]
df.groupby("BusinessTravel").mean()