Flink: Add Flink SQL procedures for Iceberg similarly to Apache Paimon

### Feature Request / Improvement

### Motivation
Apache Flink supports [CALL sql statements](https://nightlies.apache.org/flink/flink-docs-master/docs/sql/reference/utility/call/) to call a stored procedure which can perform administrative tasks. I want to implement something similar to [Paimon's procedures](https://paimon.apache.org/docs/master/flink/procedures/) in Iceberg.

The idea is to make snapshot references easier to manage directly from Flink SQL.

### Implemented procedures
- `create_tag`
- `delete_tag`
- `create_tag_from_timestamp`
- `create_branch`
- `delete_branch`
- `rollback_to`
- `rollback_to_timestamp`
- `expire_snapshots`

We can add more in the future, something like `rename_tag` or `list_snapshots` etc.

### Example usage
`CALL sys.create_tag('tableId', 'tagName', snapshotId, 'timeRetained')`

This made branch/tag workflows much more convenient in Flink SQL.

### Query engine

Flink

### Willingness to contribute

- [x] I can contribute this improvement/feature independently
- [ ] I would be willing to contribute this improvement/feature with guidance from the Iceberg community
- [ ] I cannot contribute this improvement/feature at this time

Start with Flink's CALL statement documentation and Apache Paimon's procedure documentation, then map the listed procedures to Iceberg snapshot references. The work is done when create/delete tag and branch operations, rollback operations, and snapshot expiration are available through Flink SQL, with the example CALL form supported.

Written by the indexing model from the issue text.