[Pipe] Avoid no-op ConfigNode consensus writes for covered progress

[Caideyipi]

Description

Motivation

A DataNode pipe heartbeat can report a composite progress index that is already fully covered by the ConfigNode progress. The previous isAfter() || equals() guard does not represent greater-than-or-equal semantics for partially ordered indexes such as RecoverProgressIndex. It can therefore merge to the same logical value, set the ConfigNode consensus-write flag, and periodically submit a no-op PipeHandleMetaChangeProcedure, producing unnecessary Ratis log entries.

Changes

• Use ProgressIndex.isEqualOrAfter() to skip updates when ConfigNode already covers the DataNode progress.
• Keep consensus writes when the DataNode heartbeat genuinely advances progress.
• Log the coordinator progress captured before an actual update.
• Add regression tests for both covered and advancing composite progress indexes.

Validation

• mvn spotless:apply -pl iotdb-core/confignode
• mvn -pl iotdb-core/confignode -Dtest=PipeHeartbeatParserTest -DfailIfNoTests=false test

This PR has:

• been self-reviewed.
• added unit tests to cover the new code paths.

Key changed classes

• PipeHeartbeatParser
• PipeHeartbeatParserTest