Support Filter pushdown in Spark Structured Streaming

Currently, Spark Structured streaming via the DSv2 api does not pushdown predicate. This results in more data being scan and filtered out at engine layer. This results in excessive I/O, driver bottlenecks and increased latency.

Relevant Iceberg issue - https://github.com/apache/iceberg/issues/15692

PR on Spark side - https://github.com/apache/spark/pull/55679