* add new practice exercise: hamming

[jimmytty]

I have removed all "error handling" test cases.

The difficulty was set to 8. This exercise need looping through each character from strands to compare and count differences. (my own opinion) Loop it's not an easy feature/concept for SQL, it's need a recursive cte.

However,
From all 77 active tracks, only the tracks: cobol, rust and vimscript have set exercise difficulty above "easy". Those tracks set difficulty to 4 (medium).

I can change the difficulty if you want and I can also introduce the "error handling" tests. But in this case, I'll need an opinion on whether to use the result column as json object or set invalid/errors as -1, or even something else.

[IsaacG]

Being hard where other tracks mark this as easy is fine. Other languages have loops.
I'd love some input from @exercism/sqlite on whether it makes sense to skip those tests or not. Unequal lengths feel more like part of the core problem to me for this exercise.

If we do skip the tests, we should include a comment with the skips.