[SPARK-47085][SQL][3.4] reduce the complexity of toTRowSet from n^2 to n

### What changes were proposed in this pull request?
reduce the complexity of RowSetUtils.toTRowSet from n^2 to n

### Why are the changes needed?
This causes performance issues.

### Does this PR introduce _any_ user-facing change?
No

### How was this patch tested?
Tests + test manually on AWS EMR

### Was this patch authored or co-authored using generative AI tooling?
No

Closes #45164 from igreenfield/branch-3.4.

Authored-by: Izek Greenfield <izek.greenfield@adenza.com>
Signed-off-by: Dongjoon Hyun <dhyun@apple.com>

sql/hive-thriftserver/src/main/scala/org/apache/spark/sql/hive/thriftserver/RowSetUtils.scala

@@ -52,11 +52,7 @@ object RowSetUtils {
       rows: Seq[Row],
       schema: Array[DataType],
       timeFormatters: TimeFormatters): TRowSet = {
-    var i = 0
-    val rowSize = rows.length
-    val tRows = new java.util.ArrayList[TRow](rowSize)
-    while (i < rowSize) {
-      val row = rows(i)
+    val tRows = rows.map { row =>
       val tRow = new TRow()
       var j = 0
       val columnSize = row.length
@@ -65,9 +61,8 @@ object RowSetUtils {
         tRow.addToColVals(columnValue)
         j += 1
       }
-      i += 1
-      tRows.add(tRow)
-    }
+      tRow
+    }.asJava
     new TRowSet(startRowOffSet, tRows)
   }
 
@@ -159,8 +154,7 @@ object RowSetUtils {
     val size = rows.length
     val ret = new java.util.ArrayList[T](size)
     var idx = 0
-    while (idx < size) {
-      val row = rows(idx)
+    rows.foreach { row =>
       if (row.isNullAt(ordinal)) {
         nulls.set(idx, true)
         ret.add(idx, defaultVal)

sql/hive-thriftserver/src/main/scala/org/apache/spark/sql/hive/thriftserver/SparkExecuteStatementOperation.scala

@@ -113,7 +113,7 @@ private[hive] class SparkExecuteStatementOperation(
     val offset = iter.getPosition
     val rows = iter.take(maxRows).toList
     log.debug(s"Returning result set with ${rows.length} rows from offsets " +
-      s"[${iter.getFetchStart}, ${offset}) with $statementId")
+      s"[${iter.getFetchStart}, ${iter.getPosition}) with $statementId")
     RowSetUtils.toTRowSet(offset, rows, dataTypes, getProtocolVersion, getTimeFormatters)
   }