faster checksum phase

[morgo]

Usually the checksum phase takes about 10-20% of the time that the copier phase does. However, on some very IO bound cases I have noticed it take significantly longer.

Much earlier in development we offered a separate checksum-threads concurrency setting, but unified it with threads to simplify configuration.

Because in resume-from-checkpoint we resume from the start of the checksum again, it's important that we keep the checksum fast. We could either use a static rule like double the threads used on copier + double the chunk time, or we could offer a configuration variable. TBD.