Dependencies: Change handling of exllamav2 checks

ExllamaV2 should check for solely exllamav2, otherwise errors don't
make sense. Migrate the combined "exl2" computed property to "inference"
since those are the required dependencies for minimal inference.

Signed-off-by: kingbri <bdashore3@proton.me>

common/model.py

@@ -15,7 +15,7 @@ from common.networking import handle_request_error
 from common.tabby_config import config
 from common.optional_dependencies import dependencies
 
-if dependencies.exl2:
+if dependencies.exllamav2:
     from backends.exllamav2.model import ExllamaV2Container
 
     # Global model container

common/optional_dependencies.py

@@ -25,7 +25,7 @@ class DependenciesModel(BaseModel):
 
     @computed_field
     @property
-    def exl2(self) -> bool:
+    def inference(self) -> bool:
         return self.torch and self.exllamav2 and self.flash_attn