[SPARK-4522][SQL] Parse schema with missing metadata.

This is just a quick fix for 1.2.  SPARK-4523 describes a more complete solution.

Author: Michael Armbrust <michael@databricks.com>

Closes #3392 from marmbrus/parquetMetadata and squashes the following commits:

bcc6626 [Michael Armbrust] Parse schema with missing metadata.

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/types/dataTypes.scala

@@ -84,6 +84,12 @@ object DataType {
         ("nullable", JBool(nullable)),
         ("type", dataType: JValue)) =>
       StructField(name, parseDataType(dataType), nullable, Metadata.fromJObject(metadata))
+    // Support reading schema when 'metadata' is missing.
+    case JSortedObject(
+        ("name", JString(name)),
+        ("nullable", JBool(nullable)),
+        ("type", dataType: JValue)) =>
+      StructField(name, parseDataType(dataType), nullable)
   }
 
   @deprecated("Use DataType.fromJson instead", "1.2.0")