In the context of exploratory data analysis (EDA) within Databricks, which tool or library is most appropriate for visualizing the distribution of a numerical feature across various categories?