[SPARK-23424][SQL]Add codegenStageId in comment

[kiszk]

What changes were proposed in this pull request?

This PR always adds codegenStageId in comment of the generated class. This is a replication of #20419 for post-Spark 2.3.
Closes#20419

/* 001 */ public Object generate(Object[] references) {
/* 002 */ return new GeneratedIteratorForCodegenStage1(references);
/* 003 */ }
/* 004 */
/* 005 */ // codegenStageId=1
/* 006 */ final class GeneratedIteratorForCodegenStage1 extends org.apache.spark.sql.execution.BufferedRowIterator {
/* 007 */ private Object[] references;
...

How was this patch tested?

Existing tests

[SparkQA]

Test build #87452 has finished for PR 20612 at commit 1bf0640.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[kiszk]

ping @gatorsmile, @rednaxelafx

[rednaxelafx]

LGTM. Let's wait for one more LGTM from @gatorsmile / @cloud-fan .

[cloud-fan]

although the caller should guarantee placeholderId is unique, shall we add a check here for safe? e.g. assert(!placeHolderToComments.containsKey("placeholderId"))

[kiszk]

+1, sure

[cloud-fan]

LGTM

[SparkQA]

Test build #87555 has finished for PR 20612 at commit eed9dcd.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[cloud-fan]

thanks, merging to master!