Related to #383, basically fix the same condition for the markdown and html renderers of Spark DataFrames. The fix for #383 only corrected pandas DF.