[FLINK-38489][table] Avoid unnecessary DELETE inference for ChangelogNormalize

[Au-Miner]

What is the purpose of the change

Improve the changelog mode inference for ChangelogNormalize by excluding DELETE when the operator has no filter and its input cannot produce deletions. This prevents an overly broad changelog mode from blocking planner optimizations such as delta join conversion.

Brief change log

• Update FlinkChangelogModeInferenceProgram.SatisfyModifyKindSetTraitVisitor to infer only INSERT and UPDATE modify kinds for a StreamPhysicalChangelogNormalize without a filter when its input does not contain DELETE.
• Preserve the all-changes mode when the input contains DELETE or a filter can turn an update into a deletion.
• Update changelog-inference, table-scan, and delta-join plan tests for the refined changelog mode.

Verifying this change

Added ChangelogModeInferenceTest cases verifying that:

• A ChangelogNormalize without a filter and without input deletions produces [I, UB, UA].
• A ChangelogNormalize with a filter still produces [I, UB, UA, D].

Updated TableScanTest and DeltaJoinTest expected plans to verify that the refined mode removes unnecessary ChangelogNormalize nodes and enables delta-join conversion for I, UA inputs.

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
• If yes, how is the feature documented? not applicable

[flinkbot]

CI report:

• 7e763cb Azure: SUCCESS

Bot commands

The @flinkbot bot supports the following commands:

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