[SUPPORT] split_reader don't checkpoint before consuming all splits

**split_reader don't checkpoint before consuming all splits**

- Have you gone through our [FAQs](https://hudi.apache.org/learn/faq/)?

- Join the mailing list to engage in conversations and get faster support at dev-subscribe@hudi.apache.org.

- If you have triaged this as a bug, then file an [issue](https://issues.apache.org/jira/projects/HUDI/issues) directly.

**Describe the problem you faced**

When using flink to incrementally query a mor table with many splits（read.start-commit=earliest），The first checkpoint of the reader is successful after all splits are consumed. It takes a lot of time to cause checkpoint timeout. In my understanding, the process of reading data based on mini batch can trigger and complete the checkpoint.


**To Reproduce**

Steps to reproduce the behavior:

1. mor table with many splits
2. read.start-commit=ealiest
3. read.tasks=1
4. checkpoint interval=10s and checkpoint timeout=10min

**Environment Description**

* Hudi version : 0.11.1

* flink version : 1.14.5

**Stacktrace**

`Caused by: org.apache.flink.util.FlinkRuntimeException: Exceeded checkpoint tolerable failure threshold.`