Allow quantize to only copy tensors, some other improvements (#2931)

* Allow quantize tool to only copy tensors to allow repackaging models.

* Slightly better logic when requantizing.

* Change help message to go to `stdout`.

llama.h

@@ -164,6 +164,7 @@ extern "C" {
         enum llama_ftype ftype;      // quantize to this llama_ftype
         bool allow_requantize;       // allow quantizing non-f32/f16 tensors
         bool quantize_output_tensor; // quantize output.weight
+        bool only_copy;              // only copy tensors - ftype, allow_requantize and quantize_output_tensor are ignored
     } llama_model_quantize_params;
 
     // grammar types