[ET-VK] Add CUNET model test

CUNET uses 3 operators:
```
aten.convolution.default
aten.max_pool2d_with_indices.default
aten.relu.default
```

which correspond to these signatures:
```
- func: convolution(Tensor input, Tensor weight, Tensor? bias, int[] stride, SymInt[] padding, int[] dilation, bool transposed, SymInt[] output_padding, int groups) -> Tensor
- func: max_pool2d_with_indices(Tensor self, int[2] kernel_size, int[2] stride=[], int[2] padding=0, int[2] dilation=1, bool ceil_mode=False) -> (Tensor, Tensor)
- func: relu(Tensor self) -> Tensor
```

Differential Revision: [D54700658](https://our.internmc.facebook.com/intern/diff/D54700658/)

[ghstack-poisoned]

Author: jorgep31415

backends/vulkan/test/TARGETS

@@ -20,6 +20,7 @@ python_unittest(
         "//executorch/extension/pybindings:portable_lib",  # @manual
         "//executorch/extension/pytree:pylib",
         "//executorch/kernels/portable:custom_ops_generated_lib",
+        "//wearables/camera/ml/model_experimentation/person_segmentation/cunet:cunet_lib",
     ],
 )
 

backends/vulkan/test/test_vulkan_delegate.py

@@ -17,6 +17,9 @@
 
 from executorch.exir import EdgeProgramManager, to_edge
 from torch.export import Dim, export, ExportedProgram
+from wearables.camera.ml.model_experimentation.person_segmentation.cunet.cunet import (
+    C_UNet,
+)
 
 ctypes.CDLL("libvulkan.so.1")
 
@@ -628,3 +631,13 @@ def forward(self, x):
             sample_inputs,
             memory_layouts=[vk_graph_schema.VkMemoryLayout.TENSOR_CHANNELS_PACKED],
         )
+
+    def test_vulkan_backend_cunet(self):
+        module = C_UNet()
+        sample_inputs = (torch.rand((1, 3, 96, 72), dtype=torch.float32),)
+
+        self.lower_module_and_test_output(
+            module,
+            sample_inputs,
+            memory_layouts=[vk_graph_schema.VkMemoryLayout.TENSOR_CHANNELS_PACKED],
+        )