Updates AddJar according to conventions introduced in apache#2215

Author: liancheng

sql/hive/src/main/scala/org/apache/spark/sql/hive/HiveStrategies.scala

@@ -195,8 +195,7 @@ private[hive] trait HiveStrategies {
 
   case class HiveCommandStrategy(context: HiveContext) extends Strategy {
     def apply(plan: LogicalPlan): Seq[SparkPlan] = plan match {
-      case logical.NativeCommand(sql) =>
-        NativeCommand(sql, plan.output)(context) :: Nil
+      case logical.NativeCommand(sql) => NativeCommand(sql, plan.output)(context) :: Nil
 
       case hive.DropTable(tableName, ifExists) => execution.DropTable(tableName, ifExists) :: Nil
 

sql/hive/src/main/scala/org/apache/spark/sql/hive/execution/commands.scala

@@ -66,19 +66,13 @@ case class DropTable(tableName: String, ifExists: Boolean) extends LeafNode with
  */
 @DeveloperApi
 case class AddJar(path: String) extends LeafNode with Command {
-
   def hiveContext = sqlContext.asInstanceOf[HiveContext]
 
   override def output = Seq.empty
 
-  override protected[sql] lazy val sideEffectResult: Seq[Any] = {
+  override protected[sql] lazy val sideEffectResult: Seq[Row] = {
     hiveContext.runSqlHive(s"ADD JAR $path")
     hiveContext.sparkContext.addJar(path)
-    Seq.empty[Any]
-  }
-
-  override def execute(): RDD[Row] = {
-    sideEffectResult
-    hiveContext.emptyResult
+    Seq.empty[Row]
   }
 }