[MINOR][SQL] Remove unnecessary MiMa excludes

[huaxingao]

What changes were proposed in this pull request?

I found a few unnecessary MiMa excludes when auditing SQL binary incompatible changes.

Why are the changes needed?

These MiMa excludes are not required any more, so remove.

Does this PR introduce any user-facing change?

No

How was this patch tested?

Manually tested

[HyukjinKwon]

@huaxingao, can you point out why this is not needed anymore?

[huaxingao]

I think these are not needed any more because of the following was added in #23086

    // Data Source V2 API changes
    (problem: Problem) => problem match {
      case MissingClassProblem(cls) =>
        !cls.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case MissingTypesProblem(newCls, _) =>
        !newCls.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case InheritedNewAbstractMethodProblem(cls, _) =>
        !cls.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case DirectMissingMethodProblem(meth) =>
        !meth.owner.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case ReversedMissingMethodProblem(meth) =>
        !meth.owner.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case _ => true
    }

Then, because of #25700

org.apache.spark.sql.sources.v2.reader -> org.apache.spark.sql.connector.read

I think we can actually change the above code to the following

    (problem: Problem) => problem match {
      case MissingClassProblem(cls) =>
        !cls.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case _ => true

[SparkQA]

Test build #119058 has finished for PR 27729 at commit 35af07d.

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

[huaxingao]

also cc @gatorsmile @cloud-fan

[HyukjinKwon]

Okay, LGTM

[HyukjinKwon]

Merged to master and branch-3.0.

[huaxingao]

@HyukjinKwon
You are too fast :)
I am actually thinking of making one more change for this PR. I will change

    // Data Source V2 API changes
    (problem: Problem) => problem match {
      case MissingClassProblem(cls) =>
        !cls.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case MissingTypesProblem(newCls, _) =>
        !newCls.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case InheritedNewAbstractMethodProblem(cls, _) =>
        !cls.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case DirectMissingMethodProblem(meth) =>
        !meth.owner.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case ReversedMissingMethodProblem(meth) =>
        !meth.owner.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case _ => true
    }

to

    // Data Source V2 API changes
    (problem: Problem) => problem match {
      case MissingClassProblem(cls) =>
        !cls.fullName.startsWith("org.apache.spark.sql.sources.v2")
      case _ => true

because of #25700
I guess I will have a follow-up?

[HyukjinKwon]

Sure, I think it might be best to make it as a followup of SPARK-28998.

This PR alone can be better separate anyway because these Mima exclusions in this PR were added as of SPARK-26124.