[SPARK-22847][CORE] Remove redundant code in AppStatusListener while assigning schedulingPool for stage

## What changes were proposed in this pull request?

In AppStatusListener's onStageSubmitted(event: SparkListenerStageSubmitted) method, there are duplicate code:
```
// schedulingPool was assigned twice with the same code
stage.schedulingPool = Option(event.properties).flatMap { p =>
      Option(p.getProperty("spark.scheduler.pool"))
    }.getOrElse(SparkUI.DEFAULT_POOL_NAME)
...
...
...
stage.schedulingPool = Option(event.properties).flatMap { p =>
      Option(p.getProperty("spark.scheduler.pool"))
    }.getOrElse(SparkUI.DEFAULT_POOL_NAME)

```
But, it does not make any sense to do this and there are no comment to explain for this.

## How was this patch tested?
N/A

Author: wuyi <[email protected]>

Closes apache#20033 from Ngone51/dev-spark-22847.

Author: Ngone51

core/src/main/scala/org/apache/spark/status/AppStatusListener.scala

@@ -329,10 +329,6 @@ private[spark] class AppStatusListener(
       .toSeq
     stage.jobIds = stage.jobs.map(_.jobId).toSet
 
-    stage.schedulingPool = Option(event.properties).flatMap { p =>
-      Option(p.getProperty("spark.scheduler.pool"))
-    }.getOrElse(SparkUI.DEFAULT_POOL_NAME)
-
     stage.description = Option(event.properties).flatMap { p =>
       Option(p.getProperty(SparkContext.SPARK_JOB_DESCRIPTION))
     }