Build multiple CPU variants and pick the best

This reduces the built-in linux version to not use any vector extensions
which enables the resulting builds to run under Rosetta on MacOS in
Docker.  Then at runtime it checks for the actual CPU vector
extensions and loads the best CPU library available

llm/ext_server_default.go

@@ -1,4 +1,4 @@
-//go:build !windows
+//go:build darwin
 
 package llm
 
@@ -14,6 +14,8 @@ import (
 	"github.com/jmorganca/ollama/api"
 )
 
+// TODO - explore shifting Darwin to a dynamic loading pattern for consistency with Linux and Windows
+
 type llamaExtServer struct {
 	api.Options
 }