[lake/iceberg] Support ADD COLUMN schema evolution for Iceberg lake catalog

[fresh-borzoni]

Search before asking

• I searched in the issues and found nothing similar.

Motivation

PR #2189 added ADD COLUMN support for Paimon-enabled tables, including the server-side infrastructure (syncSchemaChangesToLake, idempotency checks). However, IcebergLakeCatalog.alterTable() still throws UnsupportedOperationException for schema changes.

When a user runs ALTER TABLE ADD COLUMN on a Fluss table with Iceberg datalake enabled, the schema change is applied to Fluss but not propagated to Iceberg, breaking lake queries.

This is the Iceberg counterpart of #2128 and a sub-task of #2067.

Solution

Implement AddColumn handling in IcebergLakeCatalog.alterTable() using Iceberg's UpdateSchema API, matching Paimon's constraints and idempotency pattern:

• Use updateSchema.addColumn() + moveBefore(__bucket) to position new columns before system columns
• 3-way schema compatibility check for crash recovery (same pattern as PaimonLakeCatalog)
• Recursive type comparison ignoring Iceberg-assigned field IDs (needed because Iceberg reassigns field IDs during table creation, which breaks Type.equals() for complex types like Map/Array/Struct)
• Same constraints as Paimon: nullable columns only, LAST position only

Anything else?

No response

Willingness to contribute

• I'm willing to submit a PR!