Use ProxyArray.torch accessor in fabric view tests

wp.to_torch on a ProxyArray is deprecated in favor of the .torch
accessor.  Switch the three call sites that consume the ProxyArray
returned by get_world_poses; leave get_scales call sites alone since
that method still returns a raw wp.array (no .torch accessor).

Author: pv-nvidia

source/isaaclab_physx/test/sim/test_views_xform_prim_fabric.py

@@ -186,7 +186,7 @@ def test_fabric_set_world_does_not_write_back_to_usd(device, view_factory):
 
     # Verify Fabric has the new position
     fab_pos, _ = view.get_world_poses()
-    pos_torch = wp.to_torch(fab_pos)
+    pos_torch = fab_pos.torch
     assert torch.allclose(pos_torch, torch.tensor([[99.0, 99.0, 99.0]], device=device), atol=0.1), (
         f"Fabric should have new position, got {pos_torch}"
     )
@@ -247,7 +247,7 @@ def force_topology_changed():
     # Read back — proves the rebuilt _view_to_fabric and _fabric_world_matrices
     # are still consistent.
     ret_pos, _ = view.get_world_poses()
-    pos_torch = wp.to_torch(ret_pos)
+    pos_torch = ret_pos.torch
     expected = torch.tensor([[4.0, 5.0, 6.0], [4.0, 5.0, 6.0]], device=device)
     assert torch.allclose(pos_torch, expected, atol=1e-7), f"Read after rebuild failed on {device}: {pos_torch}"
 
@@ -273,7 +273,7 @@ def test_fabric_cuda1_world_pose_roundtrip(device, view_factory):
     view.set_world_poses(positions=new_pos)
 
     ret_pos, _ = view.get_world_poses()
-    pos_torch = wp.to_torch(ret_pos)
+    pos_torch = ret_pos.torch
     expected = torch.tensor([[10.0, 20.0, 30.0], [10.0, 20.0, 30.0]], device=device)
     assert torch.allclose(pos_torch, expected, atol=1e-7), f"Roundtrip failed on {device}: {pos_torch}"