if tdengine3 leader crash ，could not elect a new leader in K8s

Bug Description
用官方提供k8s脚本部署tdengine3.0.2.5,配置3个mnode一旦leader机子宕机,选举会有问题导致一直无法选出leader,整个集群无法工作

To Reproduce

1. 用官方k8s脚本(https://docs.taosdata.com/deployment/k8s/) 创建出replica为3的tdengine集群
2. 进入集群通过 create mnode on dnode <dnode_id>; 的方式将3个replica都任命为mnode
3. 然后在k8s下通过命令删除leader的replica(pod),后续此replica(pod)会自动重启
4. 但即便自动重启完成,通过第二台replica(pod)的taos命令运行 show mnodes; 也会发现一直无法选出leader的现象

Expected Behavior
能选举成功

Screenshots

Environment (please complete the following information):

• TDengine Version 3.0.2.5

Additional Context
原先的leader(mnode_id=1)的pod重启后的相关日志

03/10 02:04:12.273101 00000090 SYN vgId:1, begin election, sync:candidate, term:12, commit-index:10, first-ver:0, last-ver:26, min:-1, snap:10, snap-term:1, elect-times:9, as-leader-times:0, cfg-ch-times:0, hb-slow:0, hbr-slow:0, aq-items:-1, snaping:-1, replicas:3, last-cfg:-1, chging:0, restore:0, quorum:2, elect-lc-timer:10, hb:0, buffer:[10 10 26, 27), repl-mgrs:{0:0 [0 0, 0), 1:0 [0 0, 0), 2:0 [0 0, 0)}, members:{num:3, as:0, [tdengine-0.taosd.experimentb.svc.cluster.local:6030, tdengine-1.taosd.experimentb.svc.cluster.local:6030, tdengine-2.taosd.experimentb.svc.cluster.local:6030]}, hb:{0:1678413805684,1:1678413805684,2:1678413805684}, hb-reply:{0:1678413805684,1:1678413805684,2:1678413805684}
03/10 02:04:12.292439 00000090 SYN vgId:1, succeed to write raft store file:/var/lib/taos//mnode/sync/raft_store.json, term:13
03/10 02:04:12.311026 00000090 SYN vgId:1, succeed to write raft store file:/var/lib/taos//mnode/sync/raft_store.json, term:13
03/10 02:04:12.329363 00000090 SYN vgId:1, succeed to write raft store file:/var/lib/taos//mnode/sync/raft_store.json, term:13
03/10 02:04:13.545751 00000098 DND ERROR failed to send status req since Sync not leader, epSet:{tdengine-2.taosd.experimentb.svc.cluster.local:6030, tdengine-0.taosd.experimentb.svc.cluster.local:6030, tdengine-1.taosd.experimentb.svc.cluster.local:6030}, inUse:0

Start with the official Kubernetes deployment steps in the linked documentation, then reproduce the failure with three mnode replicas by deleting the leader pod. Inspect the mnode election and Raft log output, especially the term, quorum, membership, and leader-status messages. Done means the remaining replicas elect a leader and the cluster becomes usable after the pod restarts.