[SPARK-7467] Dag visualization: treat checkpoint as an RDD operation

Such that a checkpoint RDD does not go into random scopes on the UI, e.g. `take`. We've seen this in streaming.

Author: Andrew Or <andrew@databricks.com>

Closes #6004 from andrewor14/dag-viz-checkpoint and squashes the following commits:

9217439 [Andrew Or] Fix checkpoints
4ae8806 [Andrew Or] Merge branch 'master' of github.com:apache/spark into dag-viz-checkpoint
19bc07b [Andrew Or] Treat checkpoint as an RDD operation

core/src/main/scala/org/apache/spark/rdd/RDD.scala

@@ -1523,13 +1523,15 @@ abstract class RDD[T: ClassTag](
    * has completed (therefore the RDD has been materialized and potentially stored in memory).
    * doCheckpoint() is called recursively on the parent RDDs.
    */
-  private[spark] def doCheckpoint() {
-    if (!doCheckpointCalled) {
-      doCheckpointCalled = true
-      if (checkpointData.isDefined) {
-        checkpointData.get.doCheckpoint()
-      } else {
-        dependencies.foreach(_.rdd.doCheckpoint())
+  private[spark] def doCheckpoint(): Unit = {
+    RDDOperationScope.withScope(sc, "checkpoint", false, true) {
+      if (!doCheckpointCalled) {
+        doCheckpointCalled = true
+        if (checkpointData.isDefined) {
+          checkpointData.get.doCheckpoint()
+        } else {
+          dependencies.foreach(_.rdd.doCheckpoint())
+        }
       }
     }
   }