[SPARK-15707][SQL] Make Code Neat - Use map instead of if check.

## What changes were proposed in this pull request?
In forType function of object RandomDataGenerator, the code following:
if (maybeSqlTypeGenerator.isDefined){
  ....
  Some(generator)
} else{
 None
}
will be changed. Instead, maybeSqlTypeGenerator.map will be used.

## How was this patch tested?
All of the current unit tests passed.

Author: Weiqing Yang <[email protected]>

Closes apache#13448 from Sherry302/master.

sql/catalyst/src/test/scala/org/apache/spark/sql/RandomDataGenerator.scala

@@ -236,19 +236,15 @@ object RandomDataGenerator {
         // convert it to catalyst value to call udt's deserialize.
         val toCatalystType = CatalystTypeConverters.createToCatalystConverter(udt.sqlType)
 
-        if (maybeSqlTypeGenerator.isDefined) {
-          val sqlTypeGenerator = maybeSqlTypeGenerator.get
-          val generator = () => {
+        maybeSqlTypeGenerator.map { sqlTypeGenerator =>
+          () => {
             val generatedScalaValue = sqlTypeGenerator.apply()
             if (generatedScalaValue == null) {
               null
             } else {
               udt.deserialize(toCatalystType(generatedScalaValue))
             }
           }
-          Some(generator)
-        } else {
-          None
         }
       case unsupportedType => None
     }