I have a parquet files with a column g1 with schemaStructField(g1,IntegerType,true)Now I have a query with filter on g1.What's weird in the SQL viewer is that spark is loading all the rows from that file. Even though in the physical plan I can see th...