fix: move AI Toolkit to the bottom of the detection list

to avoid disrupting already-working LoRA

Author: Kevin Turner

invokeai/backend/patches/lora_conversions/formats.py

@@ -15,8 +15,6 @@
 
 
 def flux_format_from_state_dict(state_dict: dict, metadata: dict | None = None) -> FluxLoRAFormat | None:
-    if is_state_dict_likely_in_flux_aitoolkit_format(state_dict, metadata):
-        return FluxLoRAFormat.AIToolkit
     if is_state_dict_likely_in_flux_kohya_format(state_dict):
         return FluxLoRAFormat.Kohya
     elif is_state_dict_likely_in_flux_onetrainer_format(state_dict):
@@ -25,5 +23,7 @@ def flux_format_from_state_dict(state_dict: dict, metadata: dict | None = None)
         return FluxLoRAFormat.Diffusers
     elif is_state_dict_likely_flux_control(state_dict):
         return FluxLoRAFormat.Control
+    elif is_state_dict_likely_in_flux_aitoolkit_format(state_dict, metadata):
+        return FluxLoRAFormat.AIToolkit
     else:
         return None