[CH] support customized file scan which does not support transform

**Is your feature request related to a problem or challenge? Please describe what you are trying to do.**
When we use customized file scan( not support transform ) , the output spark columnar format data cannot be passed to a transform supported exec directly.

**Describe the solution you'd like**
A columnar(spark) to columnar(backend native) plan ts needed between file scan and its transform supported parent

**Describe alternatives you've considered**
Otherwise we can use a step combined of C2R and R2C instead
For C2R and R2C are already implemented

Start by reviewing the existing C2R and R2C implementations mentioned in the issue, then trace how a customized file scan connects to a transform-supported parent. The requested result is a columnar Spark-to-native plan between those operators without requiring a transform-unsupported scan.

Written by the indexing model from the issue text.