[SPARK-41206][SQL][FOLLOWUP] Make result of checkColumnNameDuplication stable to fix COLUMN_ALREADY_EXISTS check failed with Scala 2.13

### What changes were proposed in this pull request?
This pr add a sort when `columnAlreadyExistsError` will be thrown to make the result of `SchemaUtils#checkColumnNameDuplication` stable.

### Why are the changes needed?
Fix `COLUMN_ALREADY_EXISTS` check failed with Scala 2.13

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

### How was this patch tested?

- Pass GA
- Manual test:

```
dev/change-scala-version.sh 2.13
build/sbt clean "sql/testOnly org.apache.spark.sql.DataFrameSuite" -Pscala-2.13
build/sbt  "sql/testOnly org.apache.spark.sql.execution.datasources.json.JsonV1Suite" -Pscala-2.13
build/sbt  "sql/testOnly org.apache.spark.sql.execution.datasources.json.JsonV2Suite" -Pscala-2.13
build/sbt  "sql/testOnly org.apache.spark.sql.execution.datasources.json.JsonLegacyTimeParserSuite" -Pscala-2.13
```
All tests passed

Closes apache#38764 from LuciferYang/SPARK-41206.

Authored-by: yangjie01 <[email protected]>
Signed-off-by: Max Gekk <[email protected]>

Author: LuciferYang

sql/catalyst/src/main/scala/org/apache/spark/sql/util/SchemaUtils.scala

@@ -107,7 +107,7 @@ private[spark] object SchemaUtils {
     val names = if (caseSensitiveAnalysis) columnNames else columnNames.map(_.toLowerCase)
     // scalastyle:on caselocale
     if (names.distinct.length != names.length) {
-      val columnName = names.groupBy(identity).collectFirst {
+      val columnName = names.groupBy(identity).toSeq.sortBy(_._1).collectFirst {
         case (x, ys) if ys.length > 1 => x
       }.get
       throw QueryCompilationErrors.columnAlreadyExistsError(columnName)