[CORE] Avoid formatted comments from being messed by non-spotless lin…

…ters (especially IDE linters) (#7989)

backends-velox/src/main/scala/org/apache/spark/sql/execution/ColumnarCachedBatchSerializer.scala

@@ -46,7 +46,7 @@ case class CachedColumnarBatch(
     bytes: Array[Byte])
   extends CachedBatch {}
 
-// spotless:off
+// format: off
 /**
  * Feature:
  * 1. This serializer supports column pruning
@@ -75,7 +75,7 @@ case class CachedColumnarBatch(
  *   - Deserializer DefaultCachedBatch -> InternalRow (unsupport ColumnarToRow)
  *     -> Convert DefaultCachedBatch to InternalRow using vanilla Spark serializer
  */
-// spotless:on
+// format: on
 class ColumnarCachedBatchSerializer extends CachedBatchSerializer with SQLConfHelper with Logging {
   private lazy val rowBasedCachedBatchSerializer = new DefaultCachedBatchSerializer
 

gluten-substrait/src/main/scala/org/apache/gluten/extension/columnar/heuristic/ExpandFallbackPolicy.scala

@@ -31,7 +31,8 @@ import org.apache.spark.sql.execution.columnar.InMemoryTableScanExec
 import org.apache.spark.sql.execution.command.ExecutedCommandExec
 import org.apache.spark.sql.execution.exchange.Exchange
 
-// spotless:off
+
+// format: off
 /**
  * Note, this rule should only fallback to row-based plan if there is no harm.
  * The follow case should be handled carefully
@@ -64,7 +65,7 @@ import org.apache.spark.sql.execution.exchange.Exchange
  * @param isAdaptiveContext If is inside AQE
  * @param originalPlan The vanilla SparkPlan without apply gluten transform rules
  */
-// spotless:on
+// format: on
 case class ExpandFallbackPolicy(isAdaptiveContext: Boolean, originalPlan: SparkPlan)
   extends Rule[SparkPlan] {
   import ExpandFallbackPolicy._
@@ -106,14 +107,13 @@ case class ExpandFallbackPolicy(isAdaptiveContext: Boolean, originalPlan: SparkP
     transitionCost
   }
 
+  // format: off
   /**
    * When making a stage fall back, it's possible that we need a ColumnarToRow to adapt to last
    * stage's columnar output. So we need to evaluate the cost, i.e., the number of required
    * ColumnarToRow between entirely fallback stage and last stage(s). Thus, we can avoid possible
    * performance degradation caused by fallback policy.
    *
-   * spotless:off
-   *
    * Spark plan before applying fallback policy:
    *
    *        ColumnarExchange
@@ -136,9 +136,8 @@ case class ExpandFallbackPolicy(isAdaptiveContext: Boolean, originalPlan: SparkP
    *            Project
    *
    *  So by considering the cost, the fallback policy will not be applied.
-   *
-   * spotless:on
    */
+  // format: on
   private def countStageFallbackTransitionCost(plan: SparkPlan): Int = {
     var stageFallbackTransitionCost = 0
 

gluten-substrait/src/main/scala/org/apache/spark/sql/execution/GlutenExplainUtils.scala

@@ -147,15 +147,13 @@ object GlutenExplainUtils extends AdaptiveSparkPlanHelper {
     }
   }
 
-  // spotless:off
-  // scalastyle:off
+  // format: off
   /**
    * Given a input physical plan, performs the following tasks.
    *   1. Generates the explain output for the input plan excluding the subquery plans. 2. Generates
    *      the explain output for each subquery referenced in the plan.
    */
-  // scalastyle:on
-  // spotless:on
+  // format: on
   def processPlan[T <: QueryPlan[T]](
       plan: T,
       append: String => Unit,

gluten-substrait/src/main/scala/org/apache/spark/sql/execution/GlutenImplicits.scala

@@ -36,7 +36,8 @@ import org.apache.spark.sql.internal.SQLConf
 import scala.collection.mutable
 import scala.collection.mutable.ArrayBuffer
 
-// spotless:off
+
+// format: off
 /**
  * A helper class to get the Gluten fallback summary from a Spark [[Dataset]].
  *
@@ -52,7 +53,7 @@ import scala.collection.mutable.ArrayBuffer
  *   df.fallbackSummary
  * }}}
  */
-// spotless:on
+// format: on
 object GlutenImplicits {
 
   case class FallbackSummary(