Change region argument to be a tuple of integers

[andreasnoack]

This makes everything fully inferred. The dimension of the FFT is simply the length of the tuple. Not sure if we want this, but wanted to check if it could work

So writing

fff(x, (1,))
fft(x, (2, 3))

instead of

fff(x, 1)
fft(x, 2:3)

will get you fully inferred results in return

julia> @inferred(rfft(randn(3)))
2-element Vector{ComplexF64}:
    1.10905148310797 + 0.0im
 -1.7099022373745139 + 0.3511562597656166im

Closes #78

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.05%. Comparing base (2cae769) to head (1316f69).

Additional details and impacted files

@@                Coverage Diff                @@
##           an/noabstract      #91      +/-   ##
=================================================
- Coverage          98.09%   98.05%   -0.04%     
=================================================
  Files                  5        4       -1     
  Lines                419      411       -8     
=================================================
- Hits                 411      403       -8     
  Misses                 8        8              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.