Avoid to create a readsupport at write path in Data Source V2

https://github.com/apache/spark/commit/5fef6e3513d6023a837c427d183006d153c7102b introduced append mode support in Datasource V2.

The problem is, write path ended up

https://github.com/apache/spark/blob/e06da95cd9423f55cdb154a2778b0bddf7be984c/sql/core/src/main/scala/org/apache/spark/sql/DataFrameWriter.scala#L249

Retrieving the physical schema of the underlying storage is potentially expensive. Actually even worse: it looks odd that write path requires read side's schema.

This is a regression comparing to Spark 2.3.