Custom Metal Kernels from Python (#1325)

* start

* simple kernels working

* restructure

* inverse example working

* docs + fixes

* missing file

* fix imports

* address comments

* add docs + fix test

* Review comments + refactor to a single function

* update docs

* remove hashing

* fix contig bug in test

* back to a class

* trailing whitespace

* fix tests

* match c++ and python apis

* add link + make args kw_only

docs/src/python/fast.rst

@@ -12,3 +12,5 @@ Fast
   layer_norm
   rope
   scaled_dot_product_attention
+  affine_quantize
+  metal_kernel