Builders for FlatFileItemReader and FlatFileItemReaderWriter should allow using natural ordering of record

Expected Behavior

FlatFileItemReader and FlatFileItemReaderWriter builders allows to customize the field names of the target/source bean to write/read via :

• FlatFileItemReaderBuilder.DelimitedBuilder
• FlatFileItemReaderBuilder.FixedLengthBuilder
• FlatFileItemWriterBuilder.DelimitedBuilder
• FlatFileItemWriterBuilder.FormattedBuilder

As it is a common behavior to create a dedicated record bean for this, it could be useful to allow to use the natural ordering of record's bean instead of manually specify fields name as string.

Example :

FlatFileItemReader<MyRecord> reader = new FlatFileItemReaderBuilder<MyRecord>()
                .targetType(MyRecord.class)
                .resource(inFile)
                .delimited()
                .delimiter(",")
                .quoteCharacter('\"')
                .withRecordFieldOrdering()
                .build();

Context

Currently, you can workaround this by extracting the fields and pass it to names(...) :

String[] fieldsName = Arrays.stream(MyRecord.class.getRecordComponents())
                .map(RecordComponent::getName).toArray(String[]::new);

FlatFileItemReader<MyRecord> reader = new FlatFileItemReaderBuilder<MyRecord>()
                .targetType(MyRecord.class)
                .resource(inFile)
                .linesToSkip(1)
                .delimited()
                .delimiter(",")
                .quoteCharacter('\"')
                .names(fieldsName)
                .build();

Start at the four named builder entry points: FlatFileItemReaderBuilder.DelimitedBuilder, FixedLengthBuilder, and the corresponding FlatFileItemWriterBuilder builders. Review how targetType and names are handled, then verify that a withRecordFieldOrdering option can derive record component order for reading and writing. Done means each listed builder supports natural ordering with coverage for the requested record use case.