Adding permute_axes method for Tensor2D, Tensor3D, Tensor4D

[coreylowman]

Closes #108

Summary:

• Adds DevicePermute[2,3,4]
• Adds Permute[2,3,4]d (not exported)
• Adds Permute[2,3,4]dSugar

macro implementation used as opposed to nightly const generic version due to complexity (i.e. nightly const generic version was actually more complicated).

[codecov-commenter]

Codecov Report

Merging #169 (92b169c) into main (2d42a8e) will increase coverage by 0.20%.
The diff coverage is 99.31%.

@@            Coverage Diff             @@
##             main     #169      +/-   ##
==========================================
+ Coverage   93.42%   93.62%   +0.20%     
==========================================
  Files          79       81       +2     
  Lines        4043     4190     +147     
==========================================
+ Hits         3777     3923     +146     
- Misses        266      267       +1     

Impacted Files	Coverage Δ
src/devices/mod.rs	90.00% <ø> (ø)
src/devices/permute.rs	98.70% <98.70%> (ø)
src/tensor_ops/permute.rs	100.00% <100.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.