[Pipe] Preserve downstream sink error messages

[Caideyipi]

Description

Problem

When a pipe sink receives a failure from the downstream node, the returned TSStatus may contain the real receiver error in nested subStatus entries. The sink previously exposed only a generic MULTIPLE_ERROR or pipe wrapper message, so SHOW PIPES.ExceptionMessage did not identify the actual cause.

Design

The status handler recursively inspects nested sub-statuses and prefers a concrete downstream receiver error over generic aggregate or pipe wrapper messages. The selected message is preserved in the prior status and included in asynchronous retry summary exceptions. When the root cause has no message, its exception type is retained.

Stored retry errors are cleared when retry queues are drained, pipe events are discarded, or the sink closes, preventing stale failures from appearing in later status reports.

Tests

• PipeReceiverStatusHandlerTest: 14 tests, 0 failures
• Spotless check for node-commons and datanode
• git diff --check

---

This PR has:

• been self-reviewed.
• added comments explaining the why and intent of non-obvious code.
• added unit tests or modified existing tests to cover new code paths.

---

Key changed/added classes (or packages if there are too many classes) in this PR

• org.apache.iotdb.commons.pipe.receiver.PipeReceiverStatusHandler
• org.apache.iotdb.db.pipe.connector.protocol.thrift.async.IoTDBDataRegionAsyncSink
• Pipe receiver and async sink tests