Refactor: convert_hf_to_gguf.py (#17114)

* move conversion code to a dedicated conversion directory and split the files akin to the src/models architecture

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

convert_lora_to_gguf.py

@@ -22,12 +22,11 @@ if TYPE_CHECKING:
 if 'NO_LOCAL_GGUF' not in os.environ:
     sys.path.insert(1, str(Path(__file__).parent / 'gguf-py'))
 import gguf
-
-# reuse model definitions from convert_hf_to_gguf.py
-from convert_hf_to_gguf import LazyTorchTensor, ModelBase
-
 from gguf.constants import GGUFValueType
 
+# reuse model definitions from the conversion/ package
+from conversion import LazyTorchTensor, ModelBase, get_model_class
+
 logger = logging.getLogger("lora-to-gguf")
 
 
@@ -384,7 +383,7 @@ if __name__ == '__main__':
 
     with torch.inference_mode():
         try:
-            model_class = ModelBase.from_model_architecture(hparams["architectures"][0])
+            model_class = get_model_class(hparams["architectures"][0])
         except NotImplementedError:
             logger.error(f"Model {hparams['architectures'][0]} is not supported")
             sys.exit(1)