Improve Data File Duplicate Detection

[stanbrub]

Before new files are generated to the DH data directory for use in Benchmark tests, existing files are checked to ensure duplicate data is not produced. This is a big time-saver for larger scale (like 100,000,000), assuming there are a significant number of existing generator files in the data directory

• Change the file name convention so that the gen.def and gen.parquet files contain a hash of the generator definition
• Change the file list glob to use the hash

This story is not an immediate concern because of the expected high reuse of data generator files.

[stanbrub]

Updated Ids.uniqueName to allow a prefix to be provide. Used that to name files with a hash of the contents and search on the hash with a glob in the python "data file reuse detection" code