[SPARK-51322][SQL] Better error message for streaming subquery expression

[cloud-fan]

What changes were proposed in this pull request?

Today, if a user creates a subquery expression with a streaming query (using the new DataFrame subquery API, or manipulating logical plans directly), he/she will hit weird errors:

• for uncorrelated subquery expressions, Spark will invoke the batch planner to plan the subquery and hit an error because streaming plans are not recognized by the batch planner.
• for correlated subquery expressions in a batch query, it will be rewritten to joins, but the outer batch query is passed to the batch planner and we hit the same error as the previous one
• for correlated subquery expressions in a streaming query, the streaming execution does not go into subqueries to replace StreamingRelationV2 with StreamingDataSourceV2ScanRelation, and after subquery rewriting, the StreamingRelationV2 will remain and make Spark fail at runtime by StreamingRelationExec cannot be executed.

This PR proposes to check streaming subquery expression and fail earlier.

Why are the changes needed?

better error message

Does this PR introduce any user-facing change?

Yes, but it's only error message change

How was this patch tested?

new test

Was this patch authored or co-authored using generative AI tooling?

no

[cloud-fan]

cc @HeartSaVioR @viirya

[viirya]

KafkaSourceStressForDontFailOnDataLossSuite failed but looks like unrelated.

[info]   Cause: java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.UnknownTopicOrPartitionException: Failed to fetch metadata for partition failOnDataLoss-9-0 because metadata for topic `failOnDataLoss-9` could not be found

[the-sakthi]

LGTM

[HeartSaVioR]

+1