refactor(metal): move nn, io, grad, lora, optim to internal/metal

Co-Authored-By: Virgil <virgil@lethean.io>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

internal/metal/grad.go

@@ -1,6 +1,6 @@
 //go:build darwin && arm64
 
-package mlx
+package metal
 
 /*
 #include "mlx/c/mlx.h"

internal/metal/io.go

@@ -1,6 +1,6 @@
 //go:build darwin && arm64
 
-package mlx
+package metal
 
 /*
 #include <stdlib.h>

internal/metal/lora.go

@@ -1,6 +1,6 @@
 //go:build darwin && arm64
 
-package mlx
+package metal
 
 /*
 #include <stdlib.h>

internal/metal/nn.go

@@ -1,6 +1,6 @@
 //go:build darwin && arm64
 
-package mlx
+package metal
 
 // Linear is a fully-connected layer: y = x @ W.T + bias.
 // For quantized models, set Scales/Biases/GroupSize/Bits to use QuantizedMatmul.

internal/metal/optim.go

@@ -1,6 +1,6 @@
 //go:build darwin && arm64
 
-package mlx
+package metal
 
 import "math"