[Feature] Trigger savepoint automatically if the aligned snapshots in sources are created by savepoint

### Search before asking

- [X] I searched in the [issues](https://github.com/apache/incubator-paimon/issues) and found nothing similar.

### Motivation

Related to #1635 and #1486
When there are ETL jobs and Paimon Tables associated upstream and downstream, the upstream job creates a tag through savepoint, and downstream jobs consume the tag data and align it, they could automatically trigger a savepoint instead of checkpoint. For example, there are tables TableA, TableB and TableC, Job1 processes data from TableA to TableB, Job2 processes TableB to TableC
TableA --> Job1 --> TableB --> Job2 --> TableC
When Job1 save a tag named tagB1 in TableB by savepoint, Job2 will trigger a savepoint for tagB1 and save a tag named tagC1.

When we need to resume the above data processing, we can create a new Job1' and Job2' from the above savenpoints and tags in TableB and TableC to incremental recover data processing. cc @liming30

### Solution

_No response_

### Anything else?

_No response_

### Are you willing to submit a PR?

- [ ] I'm willing to submit a PR!

Start by tracing how aligned snapshots distinguish checkpoints from savepoints and how upstream tags are consumed across chained jobs. Define the behavior for propagating a savepoint from TableB to TableC, including recovery through the resulting tags; the issue names no files or tests, so the relevant implementation and validation points must first be located.

Written by the indexing model from the issue text.