wip: next ollama runner

implement llama and mllama model architectures in go using ggml (through
cgo)

ml/backend/ggml/backend_linux.go

@@ -0,0 +1,10 @@
+package ggml
+
+// #cgo CPPFLAGS: -D_GNU_SOURCE
+// #cgo LDFLAGS: -lm
+// #include "ggml-backend.h"
+import "C"
+
+func newBackend() *C.struct_ggml_backend {
+	return newCPUBackend()
+}