refactor: Remove unnecessary casts to list (#425)

Author: ValerianRey

src/torchjd/autogram/_module_hook_manager.py

@@ -163,7 +163,7 @@ def forward(
         module: nn.Module,
         *xs: Tensor,
     ) -> tuple[Tensor, ...]:
-        return tuple([x.detach() for x in xs])
+        return tuple(x.detach() for x in xs)
 
     # For Python version > 3.10, the type of `inputs` should become
     # tuple[BoolRef, TreeSpec, VJPType, PyTree, GramianAccumulator, nn.Module, *tuple[Tensor, ...]]

src/torchjd/autojac/_transform/_grad.py

@@ -51,7 +51,7 @@ def _differentiate(self, grad_outputs: Sequence[Tensor]) -> tuple[Tensor, ...]:
             return tuple()
 
         if len(self.outputs) == 0:
-            return tuple([torch.zeros_like(input) for input in self.inputs])
+            return tuple(torch.zeros_like(input) for input in self.inputs)
 
         grads = self._get_vjp(grad_outputs, self.retain_graph)
         return grads