[SPARK-14147][ML][SPARKR] SparkR predict should not output feature column

## What changes were proposed in this pull request?
SparkR does not support type of vector which is the default type of feature column in ML. R predict also does not output intermediate feature column. So SparkR ```predict``` should not output feature column. In this PR, I only fix this issue for ```naiveBayes``` and ```survreg```. ```kmeans``` has the right code route already and  ```glm``` will be fixed at SparkRWrapper refactor(#12294).

## How was this patch tested?
No new tests.

cc mengxr shivaram

Author: Yanbo Liang <ybliang8@gmail.com>

Closes #11958 from yanboliang/spark-14147.

mllib/src/main/scala/org/apache/spark/ml/r/AFTSurvivalRegressionWrapper.scala

@@ -44,7 +44,7 @@ private[r] class AFTSurvivalRegressionWrapper private (
   }
 
   def transform(dataset: Dataset[_]): DataFrame = {
-    pipeline.transform(dataset)
+    pipeline.transform(dataset).drop(aftModel.getFeaturesCol)
   }
 }
 

mllib/src/main/scala/org/apache/spark/ml/r/NaiveBayesWrapper.scala

@@ -37,7 +37,9 @@ private[r] class NaiveBayesWrapper private (
   lazy val tables: Array[Double] = naiveBayesModel.theta.toArray.map(math.exp)
 
   def transform(dataset: Dataset[_]): DataFrame = {
-    pipeline.transform(dataset).drop(PREDICTED_LABEL_INDEX_COL)
+    pipeline.transform(dataset)
+      .drop(PREDICTED_LABEL_INDEX_COL)
+      .drop(naiveBayesModel.getFeaturesCol)
   }
 }