Offload layers to GPU based on new model size estimates (#1850)

* select layers based on estimated model memory usage

* always account for scratch vram

* dont load +1 layers

* better estmation for graph alloc

* Update gpu/gpu_darwin.go

Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com>

* Update llm/llm.go

Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com>

* Update llm/llm.go

* add overhead for cuda memory

* Update llm/llm.go

Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com>

* fix build error on linux

* address comments

---------

Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com>

llm/ggml.go

@@ -78,7 +78,11 @@ type model interface {
 	ModelFamily() string
 	ModelType() string
 	FileType() string
-	NumLayers() int64
+	NumLayers() uint32
+	NumGQA() uint32
+	NumEmbed() uint32
+	NumHead() uint32
+	NumHeadKv() uint32
 }
 
 type container interface {