[Spark] Do not cast data type when not needed in DML commands (delta-…

…io#4158)

#### Which Delta project/connector is this regarding?
<!--
Please add the component selected below to the beginning of the pull
request title
For example: [Spark] Title of my pull request
-->

- [x] Spark
- [ ] Standalone
- [ ] Flink
- [ ] Kernel
- [ ] Other (fill in here)

## Description

This PR changes the logic of data type casting in DML operations, so
that when the expression has the desired output type no cast will be
performed. This is to avoid a performance issue when the expression is
slow to evaluate, such as for array functions on an array of structs,
i.e., `array_*(array<struct<...>>, ...)`.

For example, given `array_distinct(physical_ids)` inside a WHEN UPDATE
clause:
```
CASE WHEN (size(physical_ids#270, false) = 0) THEN array_except(physical_ids#1611, []) ELSE [] END
```
It will be first transformed into 
```
array_distinct(transform(physical_ids, (_, idx) -> physical_ids[idx]))
```
and then to 
```
transform(
  when(size(to_add) != 0, array_distinct(physical_ids), to_remove),
  (_, idx) ->
    when(size(to_add) != 0, array_distinct(physical_ids), to_remove)[idx]
```
which triples the amount of calculations.

After this PR the transformation will not happen again.

## How was this patch tested?

Manually inspecting the query plan.

## Does this PR introduce _any_ user-facing changes?

No.

spark/src/main/scala/org/apache/spark/sql/delta/UpdateExpressionsSupport.scala

@@ -112,7 +112,9 @@ trait UpdateExpressionsSupport extends SQLConfHelper with AnalysisHelper with De
       case Literal(nul, NullType) => Literal(nul, dataType)
       case otherExpr =>
         (fromExpression.dataType, dataType) match {
-          case (ArrayType(_: StructType, _), to @ ArrayType(toEt: StructType, toContainsNull)) =>
+          case (ArrayType(fromEt: StructType, fromNullable),
+              to @ ArrayType(toEt: StructType, toNullable))
+              if !(DataTypeUtils.sameType(fromEt, toEt) && fromNullable == toNullable) =>
             fromExpression match {
               // If fromExpression is an array function returning an array, cast the
               // underlying array first and then perform the function on the transformed array.
@@ -158,7 +160,7 @@ trait UpdateExpressionsSupport extends SQLConfHelper with AnalysisHelper with De
                   castIfNeeded(
                   GetArrayItem(fromExpression, i), toEt, castingBehavior, columnName)
                 val transformLambdaFunc = {
-                  val elementVar = NamedLambdaVariable("elementVar", toEt, toContainsNull)
+                  val elementVar = NamedLambdaVariable("elementVar", toEt, toNullable)
                   val indexVar = NamedLambdaVariable("indexVar", IntegerType, false)
                   LambdaFunction(structConverter(elementVar, indexVar), Seq(elementVar, indexVar))
                 }