[SPARK-19447] Remove remaining references to generated rows metric

## What changes were proposed in this pull request?

https://github.com/apache/spark/commit/b486ffc86d8ad6c303321dcf8514afee723f61f8 left behind references to "number of generated rows" metrics, that should have been removed.

## How was this patch tested?

Existing unit tests.

Author: Ala Luszczak <ala@databricks.com>

Closes #17939 from ala/SPARK-19447-fix.

sql/core/src/main/scala/org/apache/spark/sql/execution/basicPhysicalOperators.scala

@@ -340,8 +340,7 @@ case class RangeExec(range: org.apache.spark.sql.catalyst.plans.logical.Range)
   override val output: Seq[Attribute] = range.output
 
   override lazy val metrics = Map(
-    "numOutputRows" -> SQLMetrics.createMetric(sparkContext, "number of output rows"),
-    "numGeneratedRows" -> SQLMetrics.createMetric(sparkContext, "number of generated rows"))
+    "numOutputRows" -> SQLMetrics.createMetric(sparkContext, "number of output rows"))
 
   override lazy val canonicalized: SparkPlan = {
     RangeExec(range.canonicalized.asInstanceOf[org.apache.spark.sql.catalyst.plans.logical.Range])
@@ -354,7 +353,6 @@ case class RangeExec(range: org.apache.spark.sql.catalyst.plans.logical.Range)
 
   protected override def doProduce(ctx: CodegenContext): String = {
     val numOutput = metricTerm(ctx, "numOutputRows")
-    val numGenerated = metricTerm(ctx, "numGeneratedRows")
 
     val initTerm = ctx.freshName("initRange")
     ctx.addMutableState("boolean", initTerm, s"$initTerm = false;")