Fix ConfigNode leader transfer retries during removal

[CRZbulabula]

Description

This change makes removing the current ConfigNode leader resilient to transient Ratis leadership transfer failures.

Previously, the removal pre-check selected a single running follower and failed immediately when transferLeader returned an error. A transient election timeout could therefore abort the removal even when another healthy ConfigNode was available. The code also redirected to the requested transfer target without verifying which ConfigNode was actually elected.

This PR:

• tries running ConfigNode followers in deterministic node ID order
• retries only transient RatisRequestFailedException failures
• verifies the actual ConfigNode leader after every transfer attempt
• redirects the removal request to the actual elected leader
• bounds additional attempts by the remaining ConfigNode RPC timeout budget
• adds unit coverage for fallback, ambiguous transfer results, exhausted candidates, and non-retriable errors
• updates the ConfigNode removal integration test to explicitly remove the current leader

This fixes V2-1059.

Tests

• mvn test -pl iotdb-core/confignode -Dtest=NodeManagerTest
• mvn test-compile -DskipTests
• mvn test-compile -DskipTests -P with-zh-locale
• mvn verify -DskipUTs -Dit.test=IoTDBRemoveConfigNodeNormalIT -DfailIfNoTests=false -Dfailsafe.failIfNoSpecifiedTests=false -pl integration-test -am -PClusterIT -P with-integration-tests

[Copilot]

Pull request overview

Improves ConfigNode removal robustness by making leader-transfer handling resilient to transient Ratis transfer failures and by redirecting the request to the actual elected leader after transfer attempts.

Changes:

• Enhance NodeManager.transferLeader to try all running followers in deterministic ConfigNode ID order, retrying only transient RatisRequestFailedExceptions and confirming the actual leader after each attempt.
• Add unit tests covering transient failures, ambiguous transfer outcomes (transfer throws but leader changes), exhausted candidates, and non-retriable exceptions.
• Update the ConfigNode removal integration-test framework to explicitly remove the current ConfigNode leader.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
iotdb-core/confignode/src/main/java/org/apache/iotdb/confignode/manager/node/NodeManager.java	Adds deterministic multi-candidate leader-transfer attempts with transient-failure retry and post-transfer leader verification/redirect.
iotdb-core/confignode/src/test/java/org/apache/iotdb/confignode/manager/node/NodeManagerTest.java	New unit tests validating retry/fallback behavior and correct redirection semantics.
iotdb-core/confignode/src/main/i18n/en/org/apache/iotdb/confignode/i18n/ManagerMessages.java	Adds i18n log templates for new leader-transfer retry/verification warnings.
iotdb-core/confignode/src/main/i18n/zh/org/apache/iotdb/confignode/i18n/ManagerMessages.java	Chinese i18n equivalents for the new log templates.
integration-test/src/test/java/org/apache/iotdb/confignode/it/removeconfignode/IoTDBRemoveConfigNodeITFramework.java	Adjusts removal target selection to remove the current leader based on the environment’s leader wrapper and SHOW CONFIGNODES.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.