Fix IoTConsensus SyncLog retry handling

[jt2594838]

Description

Fix interrupted follower SyncLog handling

• Remove an interrupted request from SyncLogCacheQueue instead of applying it unconditionally.
• Preserve the thread interrupt flag and return an INTERNAL_SERVER_ERROR status with a clear message so the leader can retry.

Retry WRITE_PROCESS_ERROR

• Treat WRITE_PROCESS_ERROR as retriable in RetryUtils.needRetryForWrite.
• This covers SyncLog statuses produced by transient follower write failures such as IO failures in delete, WAL, MemTable, or TsFileProcessor paths.

Tests

• Added coverage ensuring interrupted SyncLog requests are not applied.
• Added coverage ensuring a follower WRITE_PROCESS_ERROR causes the leader to retry and eventually apply the request.
• Added direct unit coverage for the retry-status classification.
• Verified with:
  • mvn test -pl iotdb-core/consensus -am -Dtest=RetryUtilsTest,ReplicateTest#syncLogWriteProcessErrorTriggersLeaderRetryTest -Dsurefire.failIfNoSpecifiedTests=false
  • mvn test -pl iotdb-core/consensus -Dtest=ReplicateTest#syncLogInterruptedWhileWaitingTest
  • Chinese locale test compilation for the consensus module.

---

This PR has:

• been self-reviewed.
  • concurrent write
• added comments explaining the intent of the code.
• added unit tests or modified existing tests to cover new code paths.

---

Key changed/added classes

• IoTConsensusServerImpl
• RetryUtils
• ReplicateTest
• RetryUtilsTest