Parquet file with no field-ids, sorting is inaccurate and the columns are mixed up

[subkanthi]

When a Parquet file with no Iceberg field IDs and a sort column whose physical position differs from its field ID causes column names to be mixed up.