Fast RMS Norm (#862)

* fast rmsnorm

* no rms gpu

* kernel

* fix shared mem

* looped rms and donation in softmax

* Make the squaring in float32 to avoid underflow

* Fix the default StreamOrDevice for rope and rms_norm in fast

* nits

---------

Co-authored-by: Angelos Katharopoulos <a_katharopoulos@apple.com>

mlx/backend/metal/kernels/CMakeLists.txt

@@ -23,6 +23,7 @@ set(
   "gemv"
   "quantized"
   "random"
+  "rms_norm"
   "rope"
   "scan"
   "scaled_dot_product_attention"