enable metal gpu acceleration

ggml-metal.metal must be in the same directory as the ollama binary
otherwise llama.cpp will not be able to find it and load it.

1. go generate llama/llama_metal.go
2. go build .
3. ./ollama serve

llama/llama_metal.go

@@ -0,0 +1,5 @@
+//go:build metal
+package llama
+
+//go:generate cmake -S . -B build --fresh -DLLAMA_METAL=on
+//go:generate cmake --build build