Run tests with nimPreviewHashFarm on the 3 main back ends.

[c-blake]

Assuming CI tests pass (they do for me locally), this should be merged to keep them passing.

[c-blake]

..Unless, of course, we are just switching to this as the default for Nim-2.2 in which case we should reverse the define as per the comments around nimPreviewHashFarm to be like maybe d:nimStringHash2 { for (2)nd after aboriginal hashData hash with this one being the 3rd, really, though longer term it will be nim-version (2) as well }.

Either way, while the Nim CI seems robust to hash-order changing, it's pretty easy to imagine tests within the Packages set changing with a hash order change.

[github-actions]

Thanks for your hard work on this PR!
The lines below are statistics of the Nim compiler built from 8301533

Hint: mm: orc; opt: speed; options: -d:release
179014 lines; 8.498s; 664.309MiB peakmem