Failed to Analyze spark 1.6 history application

Nobody has claimed this yet.

We compiled dr-elephant (customSHSWork branch), and deployed it on our hadoop yarn cluster for test, we tried both REST api SparkFetcher and spark.fetchers.FSFetcher and failed to load recent jobs (mapreduce jobs can be successfully loaded). Look forward to any tips. Thanks!

here is the log:

```
04-11-2018 15:47:09 INFO [ForkJoinPool-3-worker-57] com.linkedin.drelephant.spark.fetchers.SparkRestClient : creating SparkApplication by calling REST API at http://myhost:23764/api/v1/applications/application_1521785315204_12497/logs to get eventlogs
04-11-2018 15:47:09 INFO [dr-el-executor-thread-0] com.linkedin.drelephant.spark.fetchers.SparkFetcher : Succeeded fetching data for application_1521785315204_12497
**04-11-2018 15:47:09 ERROR [dr-el-executor-thread-0] com.linkedin.drelephant.ElephantRunner : None.get
04-11-2018 15:47:09 ERROR [dr-el-executor-thread-0] com.linkedin.drelephant.ElephantRunner : java.util.NoSuchElementException: None.get
at scala.None$.get(Option.scala:313)
at scala.None$.get(Option.scala:311)
at** com.linkedin.drelephant.spark.heuristics.StagesWithFailedTasksHeuristic$Evaluator$$anonfun$getErrorsSeverity$1.apply(StagesWithFailedTasksHeuristic.scala:83)
at com.linkedin.drelephant.spark.heuristics.StagesWithFailedTasksHeuristic$Evaluator$$anonfun$getErrorsSeverity$1.apply(StagesWithFailedTasksHeuristic.scala:79)
at scala.collection.immutable.List.foreach(List.scala:318)
at com.linkedin.drelephant.spark.heuristics.StagesWithFailedTasksHeuristic$Evaluator.getErrorsSeverity(StagesWithFailedTasksHeuristic.scala:79)
at com.linkedin.drelephant.spark.heuristics.StagesWithFailedTasksHeuristic$Evaluator.x$1$lzycompute(StagesWithFailedTasksHeuristic.scala:143)
at com.linkedin.drelephant.spark.heuristics.StagesWithFailedTasksHeuristic$Evaluator.x$1(StagesWithFailedTasksHeuristic.scala:143)
at com.linkedin.drelephant.spark.heuristics.StagesWithFailedTasksHeuristic$Evaluator.stagesWithOOMError$lzycompute(StagesWithFailedTasksHeuristic.scala:143)
at com.linkedin.drelephant.spark.heuristics.StagesWithFailedTasksHeuristic$Evaluator.stagesWithOOMError(StagesWithFailedTasksHeuristic.scala:143)
at com.linkedin.drelephant.spark.heuristics.StagesWithFailedTasksHeuristic.apply(StagesWithFailedTasksHeuristic.scala:42)
at com.linkedin.drelephant.spark.heuristics.StagesWithFailedTasksHeuristic.apply(StagesWithFailedTasksHeuristic.scala:31)
at com.linkedin.drelephant.analysis.AnalyticJob.getAnalysis(AnalyticJob.java:265)
at com.linkedin.drelephant.ElephantRunner$ExecutorJob.run(ElephantRunner.java:175)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:748)
```

Start with spark.heuristics.StagesWithFailedTasksHeuristic.scala at line 83 and trace the None value reached after SparkFetcher reports success. Compare the REST API SparkFetcher and spark.fetchers.FSFetcher paths for the Spark 1.6 history application. Done means the application is analyzed without the reported None.get failure.

Written by the indexing model from the issue text.