[SPARK-20637][CORE] Remove mention of old RDD classes from comments

## What changes were proposed in this pull request?

A few comments around the code mention RDD classes that do not exist anymore. I'm not sure of the best way to replace these, so I've just removed them here.

## How was this patch tested?

Only changes code comments, no testing required

Author: Michael Mior <mmior@uwaterloo.ca>

Closes apache#17900 from michaelmior/remove-old-rdds.

Author: michaelmior

core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala

@@ -58,7 +58,7 @@ import org.apache.spark.util._
  * set of map output files, and another to read those files after a barrier). In the end, every
  * stage will have only shuffle dependencies on other stages, and may compute multiple operations
  * inside it. The actual pipelining of these operations happens in the RDD.compute() functions of
- * various RDDs (MappedRDD, FilteredRDD, etc).
+ * various RDDs
  *
  * In addition to coming up with a DAG of stages, the DAGScheduler also determines the preferred
  * locations to run each task on, based on the current cache status, and passes these to the

core/src/test/scala/org/apache/spark/CheckpointSuite.scala

@@ -114,7 +114,7 @@ trait RDDCheckpointTester { self: SparkFunSuite =>
    * RDDs partitions. So even if the parent RDD is checkpointed and its partitions changed,
    * the generated RDD will remember the partitions and therefore potentially the whole lineage.
    * This function should be called only those RDD whose partitions refer to parent RDD's
-   * partitions (i.e., do not call it on simple RDD like MappedRDD).
+   * partitions (i.e., do not call it on simple RDDs).
    *
    * @param op an operation to run on the RDD
    * @param reliableCheckpoint if true, use reliable checkpoints, otherwise use local checkpoints
@@ -388,7 +388,7 @@ class CheckpointSuite extends SparkFunSuite with RDDCheckpointTester with LocalS
     // the parent RDD has been checkpointed and parent partitions have been changed.
     // Note that this test is very specific to the current implementation of CartesianRDD.
     val ones = sc.makeRDD(1 to 100, 10).map(x => x)
-    checkpoint(ones, reliableCheckpoint) // checkpoint that MappedRDD
+    checkpoint(ones, reliableCheckpoint)
     val cartesian = new CartesianRDD(sc, ones, ones)
     val splitBeforeCheckpoint =
       serializeDeserialize(cartesian.partitions.head.asInstanceOf[CartesianPartition])
@@ -411,7 +411,7 @@ class CheckpointSuite extends SparkFunSuite with RDDCheckpointTester with LocalS
     // Note that this test is very specific to the current implementation of
     // CoalescedRDDPartitions.
     val ones = sc.makeRDD(1 to 100, 10).map(x => x)
-    checkpoint(ones, reliableCheckpoint) // checkpoint that MappedRDD
+    checkpoint(ones, reliableCheckpoint)
     val coalesced = new CoalescedRDD(ones, 2)
     val splitBeforeCheckpoint =
       serializeDeserialize(coalesced.partitions.head.asInstanceOf[CoalescedRDDPartition])

core/src/test/scala/org/apache/spark/scheduler/SparkListenerSuite.scala

@@ -184,7 +184,7 @@ class SparkListenerSuite extends SparkFunSuite with LocalSparkContext with Match
     sc.listenerBus.waitUntilEmpty(WAIT_TIMEOUT_MILLIS)
     listener.stageInfos.size should be {1}
     val stageInfo2 = listener.stageInfos.keys.find(_.stageId == 1).get
-    stageInfo2.rddInfos.size should be {3} // ParallelCollectionRDD, FilteredRDD, MappedRDD
+    stageInfo2.rddInfos.size should be {3}
     stageInfo2.rddInfos.forall(_.numPartitions == 4) should be {true}
     stageInfo2.rddInfos.exists(_.name == "Deux") should be {true}
     listener.stageInfos.clear()