[SPARK-19777] Scan runningTasksSet when check speculatable tasks in TaskSetManager.

## What changes were proposed in this pull request?

When check speculatable tasks in `TaskSetManager`, only scan `runningTasksSet` instead of scanning all `taskInfos`.

## How was this patch tested?
Existing tests.

Author: jinxing <jinxing6042@126.com>

Closes #17111 from jinxing64/SPARK-19777.

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

@@ -906,8 +906,6 @@ private[spark] class TaskSetManager(
    * Check for tasks to be speculated and return true if there are any. This is called periodically
    * by the TaskScheduler.
    *
-   * TODO: To make this scale to large jobs, we need to maintain a list of running tasks, so that
-   * we don't scan the whole task set. It might also help to make this sorted by launch time.
    */
   override def checkSpeculatableTasks(minTimeToSpeculation: Int): Boolean = {
     // Can't speculate if we only have one task, and no need to speculate if the task set is a
@@ -927,7 +925,8 @@ private[spark] class TaskSetManager(
       // TODO: Threshold should also look at standard deviation of task durations and have a lower
       // bound based on that.
       logDebug("Task length threshold for speculation: " + threshold)
-      for ((tid, info) <- taskInfos) {
+      for (tid <- runningTasksSet) {
+        val info = taskInfos(tid)
         val index = info.index
         if (!successful(index) && copiesRunning(index) == 1 && info.timeRunning(time) > threshold &&
           !speculatableTasks.contains(index)) {