[SPARK-58936][SQL] Add partial view updates to ViewCatalog

[manuzhang]

What changes were proposed in this pull request?

This PR adds a partial-update API for catalog views:

• Add ViewChange with ordered SetProperty and RemoveProperty changes.
• Add ViewCatalog.alterView(Identifier, ViewChange...).
• Provide a backward-compatible default that loads the latest view, applies the changes while preserving all typed view metadata, and calls replaceView.
• Route ALTER VIEW ... SET TBLPROPERTIES and ALTER VIEW ... UNSET TBLPROPERTIES through one alterView change batch.
• Keep full metadata operations such as ALTER VIEW ... AS and schema-binding changes on replaceView.

Catalogs can override alterView to translate the full ordered batch into a native atomic update.

Why are the changes needed?

The current V2 property ALTER executors rebuild a complete View from metadata captured during analysis and call replaceView. That prevents connectors from applying native property validation and atomic partial updates, and a full replacement can overwrite unrelated metadata changed concurrently.

A ViewChange API, analogous to TableChange, gives connectors the requested operation instead of a reconstructed full view. The default implementation preserves compatibility for existing ViewCatalog implementations.

Does this PR introduce any user-facing change?

Yes. This adds an evolving connector API for partial view updates. Existing catalogs remain compatible through the default implementation, and SQL SET/UNSET behavior is unchanged. Catalogs that override alterView can provide native validation and atomic update semantics.

How was this patch tested?

• build/sbt catalyst/Test/compile
• build/sbt sql/Test/compile
• build/sbt 'catalyst/testOnly org.apache.spark.sql.connector.catalog.ViewCatalogSuite'
• SPARK_LOCAL_IP=127.0.0.1 SPARK_LOCAL_HOSTNAME=localhost build/sbt 'sql/testOnly org.apache.spark.sql.execution.command.v2.AlterViewSetTblPropertiesSuite org.apache.spark.sql.execution.command.v2.AlterViewUnsetTblPropertiesSuite'
• build/sbt catalyst/checkstyle catalyst/scalastyle catalyst/Test/scalastyle sql/scalastyle sql/Test/scalastyle

Was this patch authored or co-authored using generative AI tooling?

Generated-by: Codex (GPT-5)

[manuzhang]

@szehon-ho @aokolnychyi @cloud-fan This is the feature gap we found when implementing Iceberg integration with Spark 4.2