Error when serializing scala spark DataFrame that comes out of a sqlContext.read.*

[lbustelo]

See https://gist.github.com/lbustelo/42f074827738ba45c0de for example.

It seems that when creating DataFrames using the sqlContext.read.* method we get an error with serialization.

/cc @vinomaster @wellecks

[lbustelo]

The problem has to do with our serialization of Dataframes. We are running
``df.take(limit).map(row => row.toSeq.toArray.map(_.toString))`

But if the value is null (not present) we get an NPE.