[SPARK-18608][ML] Fix double caching

## What changes were proposed in this pull request?
`df.rdd.getStorageLevel` => `df.storageLevel`

using cmd `find . -name '*.scala' | xargs -i bash -c 'egrep -in "\.rdd\.getStorageLevel" {} && echo {}'` to make sure all algs involved in this issue are fixed.

Previous discussion in other PRs: apache#19107, apache#17014

## How was this patch tested?
existing tests

Author: Zheng RuiFeng <[email protected]>

Closes apache#19197 from zhengruifeng/double_caching.

Author: zhengruifeng

mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala

@@ -484,7 +484,7 @@ class LogisticRegression @Since("1.2.0") (
   }
 
   override protected[spark] def train(dataset: Dataset[_]): LogisticRegressionModel = {
-    val handlePersistence = dataset.rdd.getStorageLevel == StorageLevel.NONE
+    val handlePersistence = dataset.storageLevel == StorageLevel.NONE
     train(dataset, handlePersistence)
   }
 

mllib/src/main/scala/org/apache/spark/ml/classification/OneVsRest.scala

@@ -165,7 +165,7 @@ final class OneVsRestModel private[ml] (
     val newDataset = dataset.withColumn(accColName, initUDF())
 
     // persist if underlying dataset is not persistent.
-    val handlePersistence = dataset.rdd.getStorageLevel == StorageLevel.NONE
+    val handlePersistence = dataset.storageLevel == StorageLevel.NONE
     if (handlePersistence) {
       newDataset.persist(StorageLevel.MEMORY_AND_DISK)
     }
@@ -358,7 +358,7 @@ final class OneVsRest @Since("1.4.0") (
     }
 
     // persist if underlying dataset is not persistent.
-    val handlePersistence = dataset.rdd.getStorageLevel == StorageLevel.NONE
+    val handlePersistence = dataset.storageLevel == StorageLevel.NONE
     if (handlePersistence) {
       multiclassLabeled.persist(StorageLevel.MEMORY_AND_DISK)
     }

mllib/src/main/scala/org/apache/spark/ml/clustering/KMeans.scala

@@ -304,7 +304,7 @@ class KMeans @Since("1.5.0") (
   override def fit(dataset: Dataset[_]): KMeansModel = {
     transformSchema(dataset.schema, logging = true)
 
-    val handlePersistence = dataset.rdd.getStorageLevel == StorageLevel.NONE
+    val handlePersistence = dataset.storageLevel == StorageLevel.NONE
     val instances: RDD[OldVector] = dataset.select(col($(featuresCol))).rdd.map {
       case Row(point: Vector) => OldVectors.fromML(point)
     }

mllib/src/main/scala/org/apache/spark/ml/regression/AFTSurvivalRegression.scala

@@ -213,7 +213,7 @@ class AFTSurvivalRegression @Since("1.6.0") (@Since("1.6.0") override val uid: S
   override def fit(dataset: Dataset[_]): AFTSurvivalRegressionModel = {
     transformSchema(dataset.schema, logging = true)
     val instances = extractAFTPoints(dataset)
-    val handlePersistence = dataset.rdd.getStorageLevel == StorageLevel.NONE
+    val handlePersistence = dataset.storageLevel == StorageLevel.NONE
     if (handlePersistence) instances.persist(StorageLevel.MEMORY_AND_DISK)
 
     val featuresSummarizer = {

mllib/src/main/scala/org/apache/spark/ml/regression/IsotonicRegression.scala

@@ -165,7 +165,7 @@ class IsotonicRegression @Since("1.5.0") (@Since("1.5.0") override val uid: Stri
     transformSchema(dataset.schema, logging = true)
     // Extract columns from data.  If dataset is persisted, do not persist oldDataset.
     val instances = extractWeightedLabeledPoints(dataset)
-    val handlePersistence = dataset.rdd.getStorageLevel == StorageLevel.NONE
+    val handlePersistence = dataset.storageLevel == StorageLevel.NONE
     if (handlePersistence) instances.persist(StorageLevel.MEMORY_AND_DISK)
 
     val instr = Instrumentation.create(this, dataset)

mllib/src/main/scala/org/apache/spark/ml/regression/LinearRegression.scala

@@ -251,7 +251,7 @@ class LinearRegression @Since("1.3.0") (@Since("1.3.0") override val uid: String
       return lrModel
     }
 
-    val handlePersistence = dataset.rdd.getStorageLevel == StorageLevel.NONE
+    val handlePersistence = dataset.storageLevel == StorageLevel.NONE
     if (handlePersistence) instances.persist(StorageLevel.MEMORY_AND_DISK)
 
     val (featuresSummarizer, ySummarizer) = {