fix(pipe): reuse safe local progress on alter

[Caideyipi]

Summary:

• Reuse a local committed PipeTask progress index during alter only when the old and new realtime-only modes are unchanged, the leader remains on this DataNode, and the old task is actually local.
• Preserve ConfigNode progress monotonically and retain MinimumProgressIndex reset semantics for mode changes and leader migration.
• Process alter's dropped and updated PipeMeta entries in one DataNode task-agent invocation, with timeout handling.

Tests:

• mvn spotless:check -pl iotdb-core/datanode
• 14 targeted JUnit tests passed, covering local progress, ConfigNode progress precedence, mode reset, both leader-migration directions, missing local tasks, batch handling, timeout handling, and existing multi-pipe error behavior.
• Full Maven compile remains blocked by this workspace's existing memory pressure and generated Thrift API mismatch.

[jt2594838]

Is it possible to add an IT that tests altering an pipe does not re-send already received data?

[Caideyipi]

Added testAlterPipeDoesNotResendCommittedData in 5fb019b. It transfers two rows, deletes them on the receiver, alters the sink configuration, asserts that the deleted rows are not sent again, then writes a third row and verifies that replication remains active. The sender heartbeat interval is set to 600 seconds so the test covers stale ConfigNode metadata instead of relying on a periodic heartbeat. The targeted IT passes locally.