hash without pyspark

[bjuergens]

step 1: simulate dirhash (preferable via bash) for a single small file. So basically just run SHA1 on a file (and maybe on the name, too)

then maybe create version of dirhash, that doesn't need spark. Or make a shell-script, that creates the same output, like dirhash would.

This script can be used to clearify what dirhash actually does (so we can use this script in the readme). And we could also use this script when dirhash is not available (e.g. for local testing)