[jin-sir]

Summary

• Support Iceberg DDL/DML extensions: partition evolution, write distribution/ordering, identifier fields, branch/tag, and CALL procedures.
• Support multi-level table identifiers (catalog.db.analytics.table) across CREATE / ALTER / DROP.
• Fix a base-Spark regression: unquoted dotted column declarations are rejected again, while nested Iceberg ADD COLUMN keeps entity and completion coverage.
• Collect both target and source table entities for MERGE.
• Add table suggestions for metadata-table prefixes and function suggestions for CALL.
• Restore invalid-SQL diagnostics for nested ADD COLUMN.

Verification

• Spark: 35 suites / 766 tests
• Full repository: 227 suites / 6683 tests
• check-types, build, and prettier-check all pass

Closes DTStack#463

🤖 Generated with Claude Code

[jin-sir]

Superseded by upstream PR DTStack#505