The main changes in this PR are as follows:

1. Add `topologyRefreshEnabled` and `topologyRefreshPeriod` to control the periodic topology refresh mechanism.
2. `topologyRefreshExecutor` is a single-threaded executor responsible for running `TopologyRefreshTask`.
3. Add `checkClusterSlotSequence` to check whether the cluster slots returned by the server are consecutive and there are no duplicates.
4. [Test] In JedisClusterTestBase, `CLUSTER RESET SOFT` was changed to `HARD`, because SOFT cannot clean up the Redis Cluster configuration and may cause a crash. Please refer to redis/redis#12701
5. [Test] Adjust the cluster-node-timeout in the Makefile to 15s, consistent with the default configuration of Redis.

---------

* Introducing periodic topology mechanism for JedisCluster

* Apply suggestions from sazzad16

Co-authored-by: M Sazzadul Hoque <[email protected]>

* Remove topologyRefreshEnabled

* Apply suggestions from sazzad16

Co-authored-by: M Sazzadul Hoque <[email protected]>

* remove save topologyRefreshPeriod

* Move INIT_NO_ERROR_PROPERTY to JedisCluster

* add timeout for clusterPeriodTopologyRefreshTest

* Update src/main/java/redis/clients/jedis/DefaultJedisSocketFactory.java

Co-authored-by: M Sazzadul Hoque <[email protected]>

---------

Co-authored-by: M Sazzadul Hoque <[email protected]>