The write operation currently generates a COPY command like the following:

COPY "PUBLIC"."some_table" FROM 's3://some-bucket/tmp/manifest.json' CREDENTIALS 'aws_access_key_id=__;aws_secret_access_key=__' FORMAT AS CSV NULL AS '@NULL@' manifest 

This relies on the DataFrame to have the columns in the same order as the table if it already exists. However, the COPY command supports specifying column lists or JSONPath expressions to map columns (documentation). It would be nice to at least support the column list, potentially as an option on the write operation like:

df.write
  .format("com.databricks.spark.redshift")
  .option("url", "jdbc:redshift://redshifthost:5439/database?user=username&password=pass")
  .option("dbtable", "my_table_copy")
  .option("tempdir", "s3n://path/for/temp/data")
  .option("include_column_list", "true")
  .mode("error")
  .save()

Looks like this should be fairly straightforward to add here.