Always dynamically load the llm server library

This switches darwin to dynamic loading, and refactors the code now that no
static linking of the library is used on any platform

llm/payload_linux.go

@@ -0,0 +1,8 @@
+package llm
+
+import (
+	"embed"
+)
+
+//go:embed llama.cpp/build/linux/*/lib/*.so
+var libEmbed embed.FS