[SPARK-14717] [PYTHON] Scala, Python APIs for Dataset.unpersist differ in default blocking value

## What changes were proposed in this pull request?

Change unpersist blocking parameter default value to match Scala

## How was this patch tested?

unit tests, manual tests

jkbradley davies

Author: felixcheung <felixcheung_m@hotmail.com>

Closes #12507 from felixcheung/pyunpersist.

python/pyspark/sql/dataframe.py

@@ -326,9 +326,11 @@ class DataFrame(object):
         return self
 
     @since(1.3)
-    def unpersist(self, blocking=True):
+    def unpersist(self, blocking=False):
         """Marks the :class:`DataFrame` as non-persistent, and remove all blocks for it from
         memory and disk.
+
+        .. note:: `blocking` default has changed to False to match Scala in 2.0.
         """
         self.is_cached = False
         self._jdf.unpersist(blocking)

python/pyspark/sql/tests.py

@@ -362,7 +362,7 @@ class SQLTests(ReusedPySparkTestCase):
         # cache and checkpoint
         self.assertFalse(df.is_cached)
         df.persist()
-        df.unpersist()
+        df.unpersist(True)
         df.cache()
         self.assertTrue(df.is_cached)
         self.assertEqual(2, df.count())