[SPARK-8355] [SQL] Python DataFrameReader/Writer should mirror Scala

I compared PySpark DataFrameReader/Writer against Scala ones. `Option` function is missing in both reader and writer, but the rest seems to all match.

I added `Option` to reader and writer and updated the `pyspark-sql` test.

Author: Cheolsoo Park <cheolsoop@netflix.com>

Closes #7078 from piaozhexiu/SPARK-8355 and squashes the following commits:

c63d419 [Cheolsoo Park] Fix version
524e0aa [Cheolsoo Park] Add option function to df reader and writer

python/pyspark/sql/readwriter.py

@@ -73,6 +73,13 @@ class DataFrameReader(object):
         self._jreader = self._jreader.schema(jschema)
         return self
 
+    @since(1.5)
+    def option(self, key, value):
+        """Adds an input option for the underlying data source.
+        """
+        self._jreader = self._jreader.option(key, value)
+        return self
+
     @since(1.4)
     def options(self, **options):
         """Adds input options for the underlying data source.
@@ -235,6 +242,13 @@ class DataFrameWriter(object):
         self._jwrite = self._jwrite.format(source)
         return self
 
+    @since(1.5)
+    def option(self, key, value):
+        """Adds an output option for the underlying data source.
+        """
+        self._jwrite = self._jwrite.option(key, value)
+        return self
+
     @since(1.4)
     def options(self, **options):
         """Adds output options for the underlying data source.

python/pyspark/sql/tests.py

@@ -564,6 +564,7 @@ class SQLTests(ReusedPySparkTestCase):
         self.assertEqual(sorted(df.collect()), sorted(actual.collect()))
 
         df.write.mode("overwrite").options(noUse="this options will not be used in save.")\
+                .option("noUse", "this option will not be used in save.")\
                 .format("json").save(path=tmpPath)
         actual =\
             self.sqlCtx.read.format("json")\