[SPARK-16963][SQL] Fix test "StreamExecution metadata garbage collection"

## What changes were proposed in this pull request?

A follow up PR for #14553 to fix the flaky test. It's flaky because the file list API doesn't guarantee any order of the return list.

## How was this patch tested?

Jenkins

Author: Shixiong Zhu <shixiong@databricks.com>

Closes #15661 from zsxwing/fix-StreamingQuerySuite.

sql/core/src/test/scala/org/apache/spark/sql/streaming/StreamingQuerySuite.scala

@@ -265,7 +265,7 @@ class StreamingQuerySuite extends StreamTest with BeforeAndAfter with Logging {
       AssertOnQuery("metadata log should contain only two files") { q =>
         val metadataLogDir = new java.io.File(q.offsetLog.metadataPath.toString)
         val logFileNames = metadataLogDir.listFiles().toSeq.map(_.getName())
-        val toTest = logFileNames.filter(! _.endsWith(".crc"))  // Workaround for SPARK-17475
+        val toTest = logFileNames.filter(! _.endsWith(".crc")).sorted  // Workaround for SPARK-17475
         assert(toTest.size == 2 && toTest.head == "1")
         true
       }