Feature complete Metal FFT (#1102)

* feature complete metal fft

* fix contiguity bug

* jit fft

* simplify rader/bluestein constant computation

* remove kernel/utils.h dep

* remove bf16.h dep

* format

---------

Co-authored-by: Alex Barron <abarron22@apple.com>

mlx/backend/metal/jit/includes.h

@@ -17,6 +17,7 @@ const char* unary();
 const char* binary();
 const char* binary_two();
 const char* copy();
+const char* fft();
 const char* ternary();
 const char* scan();
 const char* softmax();