[SPARK-15417][SQL][PYTHON] PySpark shell always uses in-memory catalog

## What changes were proposed in this pull request?

There is no way to use the Hive catalog in `pyspark-shell`. This is because we used to create a `SparkContext` before calling `SparkSession.enableHiveSupport().getOrCreate()`, which just gets the existing `SparkContext` instead of creating a new one. As a result, `spark.sql.catalogImplementation` was never propagated.

## How was this patch tested?

Manual.

Author: Andrew Or <andrew@databricks.com>

Closes #13203 from andrewor14/fix-pyspark-shell.

python/pyspark/shell.py

@@ -35,12 +35,11 @@ from pyspark.storagelevel import StorageLevel
 if os.environ.get("SPARK_EXECUTOR_URI"):
     SparkContext.setSystemProperty("spark.executor.uri", os.environ["SPARK_EXECUTOR_URI"])
 
-sc = SparkContext()
-atexit.register(lambda: sc.stop())
+SparkContext._ensure_initialized()
 
 try:
     # Try to access HiveConf, it will raise exception if Hive is not added
-    sc._jvm.org.apache.hadoop.hive.conf.HiveConf()
+    SparkContext._jvm.org.apache.hadoop.hive.conf.HiveConf()
     spark = SparkSession.builder\
         .enableHiveSupport()\
         .getOrCreate()
@@ -49,6 +48,9 @@ except py4j.protocol.Py4JError:
 except TypeError:
     spark = SparkSession(sc)
 
+sc = spark.sparkContext
+atexit.register(lambda: sc.stop())
+
 # for compatibility
 sqlContext = spark._wrapped
 sqlCtx = sqlContext

python/pyspark/sql/session.py

@@ -202,6 +202,12 @@ class SparkSession(object):
         """
         return self.__class__(self._sc, self._jsparkSession.newSession())
 
+    @property
+    @since(2.0)
+    def sparkContext(self):
+        """Returns the underlying :class:`SparkContext`."""
+        return self._sc
+
     @property
     @since(2.0)
     def conf(self):