feat: added to_grayscale paddle function

ivy/functional/frontends/paddle/vision/transforms.py

@@ -200,6 +200,19 @@ def pad(img, padding, fill=0, padding_mode="constant"):
         raise "Unsupported padding_mode"
 
 
+@with_supported_dtypes(
+    {"2.5.1 and below": ("float32", "float64", "int32", "int64")}, "paddle"
+)
+@to_ivy_arrays_and_back
+def to_grayscale(img, num_output_channels=1):
+    if num_output_channels == 1:
+        return ivy.mean(img, axis=-3, keepdims=True)
+    elif num_output_channels == 3:
+        return ivy.repeat(ivy.mean(img, axis=-3, keepdims=True), 3, axis=-3)
+    else:
+        raise ValueError("num_output_channels should be either 1 or 3")
+
+
 @with_supported_dtypes(
     {"2.5.1 and below": ("float32", "float64", "int32", "int64")}, "paddle"
 )

ivy_tests/test_ivy/test_frontends/test_paddle/test_vision/test_transforms.py

@@ -180,6 +180,41 @@ def test_paddle_pad(
     )
 
 
+# to_grayscale
+@handle_frontend_test(
+    fn_tree="paddle.vision.transforms.to_grayscale",
+    dtype_and_x=helpers.dtype_and_values(
+        available_dtypes=helpers.get_dtypes("valid"),
+        min_num_dims=3,
+        max_num_dims=3,
+        min_dim_size=3,
+        max_dim_size=3,
+    ),
+    num_output_channels=st.sampled_from([1, 3]),
+)
+def test_paddle_to_grayscale(
+    *,
+    dtype_and_x,
+    num_output_channels,
+    on_device,
+    fn_tree,
+    frontend,
+    test_flags,
+    backend_fw,
+):
+    input_dtype, x = dtype_and_x
+    helpers.test_frontend_function(
+        input_dtypes=input_dtype,
+        backend_to_test=backend_fw,
+        frontend=frontend,
+        test_flags=test_flags,
+        fn_tree=fn_tree,
+        on_device=on_device,
+        img=x[0],
+        num_output_channels=num_output_channels,
+    )
+
+
 # to_tensor
 @handle_frontend_test(
     fn_tree="paddle.to_tensor",