[spark] partition filter pushdown for non-lake log and kv tables

[fresh-borzoni]

Search before asking

• I searched in the issues and found nothing similar.

Description

When a Spark query on a partitioned non-lake table has a WHERE clause referencing partition columns, the connector shall extracts a partition-only sub-predicate and prune the partition list at planning time.

Partitions that can't match are skipped - no per-partition log or kv-snapshot reads are issued for them.

Willingness to contribute

• I'm willing to submit a PR!