test: add partitioned schema evolution coverage

[liujiwen-up]

Purpose

Linked issue: #258

Add integration coverage for partitioned schema evolution across mixed file formats. The fixture writes old-schema partitioned Parquet data, adds an extra column, then writes new-schema ORC and Avro data across multiple dt partitions.

Brief change log

• Add partitioned_format_schema_evolution_add_column Spark fixture with dt STRING partitioning.
• Add core read coverage for partition pruning, projection plus filter on partition/data columns, and extra IS NULL with a partition filter.
• Add DataFusion SQL coverage for partition filters, added-column null fill, and projected partition/data columns.

Tests

• python3 -m py_compile dev/spark/provision.py: passed
• cargo fmt --check: passed
• cargo test -p paimon-integration-tests --test read_tables partitioned_format_schema_evolution -- --list: passed
• cargo test -p paimon-datafusion --test read_tables partitioned_format_schema_evolution -- --list: passed
• cargo test -p paimon-integration-tests --test read_tables partitioned_format_schema_evolution --no-run: passed
• cargo test -p paimon-datafusion --test read_tables partitioned_format_schema_evolution --no-run: passed
• cargo test -p paimon-integration-tests --test read_tables partitioned_format_schema_evolution -- --nocapture: passed after re-provisioning the Spark warehouse
• cargo test -p paimon-datafusion --test read_tables partitioned_format_schema_evolution -- --nocapture: passed after re-provisioning the Spark warehouse

API and Format

No API or storage format changes. This PR only adds Spark fixture data and integration test coverage.

Documentation

No documentation update required; this is test-only coverage.

[QuakeWang]

+1