[FLINK-38960][doc] Clarify scope of state compatibility in "Upgrading…

[AlexYinHan]

What is the purpose of the change

This PR updates the "upgrading" documentation(link). In this doc, there is a table indicating the savepoint compatibility among various Flink versions. We should clarify that 'compatibility' refers specifically to compatibility of the internal data format used in savepoints and does not cover compatibility of SQL operators or other Upper-level changes. This helps to avoid user confusion.

Brief change log

This PR updates the "upgrading" documentation to clarify scope of state compatibility.

Verifying this change

This change is a trivial rework / code cleanup without any test coverage.

Does this pull request potentially affect one of the following parts:

• Dependencies (does it add or upgrade a dependency): (no)
• The public API, i.e., is any changed class annotated with @Public(Evolving): (no)
• The serializers: (no)
• The runtime per-record code paths (performance sensitive): (no)
• Anything that affects deployment or recovery: JobManager (and its components), Checkpointing, Kubernetes/Yarn, ZooKeeper: (no)
• The S3 file system connector: (no)

Documentation

• Does this pull request introduce a new feature? (no)

[flinkbot]

CI report:

• cde0230 Azure: FAILURE

Bot commands

The @flinkbot bot supports the following commands:

• @flinkbot run azure re-run the last Azure build

[davidradl]

The Jira indicates "This helps to avoid user confusion.". I am not sure how to action "cover compatibility of SQL operators or other Upper-level changes". It leads me to ask the questions

• how do I check the compatibility of SQL operators
• how can I check the other Upper-level changes

If I cannot do this, how will I know if my savepoint can be upgraded.
I wonder if we can as a minimum either:

• ask the user to investigate the code base with some guidance around their save point. I think this would help as it would be a way forward for them. Ideally we would document the criteria when the savepoints are compatible and when they are not.
• give an example of an incompatible savepoint in each of these cases where the internal data structure is compatible.