[SPARK-15869][STREAMING] Fix a potential NPE in StreamingJobProgressListener.getBatchUIData

## What changes were proposed in this pull request?

Moved `asScala` to a `map` to avoid NPE.

## How was this patch tested?

Existing unit tests.

Author: Shixiong Zhu <shixiong@databricks.com>

Closes #14443 from zsxwing/SPARK-15869.

streaming/src/main/scala/org/apache/spark/streaming/ui/StreamingJobProgressListener.scala

@@ -259,7 +259,7 @@ private[streaming] class StreamingJobProgressListener(ssc: StreamingContext)
       // We use an Iterable rather than explicitly converting to a seq so that updates
       // will propagate
       val outputOpIdToSparkJobIds: Iterable[OutputOpIdAndSparkJobId] =
-        Option(batchTimeToOutputOpIdSparkJobIdPair.get(batchTime).asScala)
+        Option(batchTimeToOutputOpIdSparkJobIdPair.get(batchTime)).map(_.asScala)
           .getOrElse(Seq.empty)
       _batchUIData.outputOpIdSparkJobIdPairs = outputOpIdToSparkJobIds
     }