SortOrderId is null after planFile operation

[rice668]

Apache Iceberg version

main (development)

Query engine

Spark

Please describe the bug 🐞

Reproduction Steps (pseudo test code)

 createTable("id bigint NOT NULL, category string");
    Table table = validationCatalog.loadTable(tableIdent);
    table.replaceSortOrder().asc("id", NullOrder.NULLS_FIRST).commit();
    DataFile dataFile = DataFiles.builder(table.spec())
        .withPath(table.location())
        .withFormat(FileFormat.ORC)
        .withRecordCount(100)
        .withFileSizeInBytes(table.newScan().targetSplitSize() * 10)
        .withSortOrder(SortOrder.builderFor(table.schema()).asc("id", NullOrder.NULLS_FIRST).build()).build();
    table.newAppend().appendFile(dataFile).commit();
    for (FileScanTask scanTask : table.newScan().planFiles()) {
          scanTask.file().sortOrderId() is null  // Sort order id information is lost
    }

[Fokko]

That doesn't look right. @zhangminglei are you interested in providing a PR?

[rice668]

Yes. I am interested.