what route did you take? Were you able to use Delta Lake/Parquet files without using a spark pool and a synapse workspace?