[SPARK-21283][CORE] FileOutputStream should be created as append mode

## What changes were proposed in this pull request?

`FileAppender` is used to write `stderr` and `stdout` files  in `ExecutorRunner`, But before writing `ErrorStream` into the the `stderr` file, the header information has been written into ,if  FileOutputStream is  not created as append mode, the  header information will be lost

## How was this patch tested?
unit test case

Author: liuxian <liu.xian3@zte.com.cn>

Closes #18507 from 10110346/wip-lx-0703.

core/src/main/scala/org/apache/spark/util/logging/FileAppender.scala

@@ -94,7 +94,7 @@ private[spark] class FileAppender(inputStream: InputStream, file: File, bufferSi
 
   /** Open the file output stream */
   protected def openFile() {
-    outputStream = new FileOutputStream(file, false)
+    outputStream = new FileOutputStream(file, true)
     logDebug(s"Opened file $file")
   }
 

core/src/test/scala/org/apache/spark/util/FileAppenderSuite.scala

@@ -52,10 +52,13 @@ class FileAppenderSuite extends SparkFunSuite with BeforeAndAfter with Logging {
   test("basic file appender") {
     val testString = (1 to 1000).mkString(", ")
     val inputStream = new ByteArrayInputStream(testString.getBytes(StandardCharsets.UTF_8))
+    // The `header` should not be covered
+    val header = "Add header"
+    Files.write(header, testFile, StandardCharsets.UTF_8)
     val appender = new FileAppender(inputStream, testFile)
     inputStream.close()
     appender.awaitTermination()
-    assert(Files.toString(testFile, StandardCharsets.UTF_8) === testString)
+    assert(Files.toString(testFile, StandardCharsets.UTF_8) === header + testString)
   }
 
   test("rolling file appender - time-based rolling") {