feat: Add __float__ method to Paddle Tensor class (#27346)

Co-authored-by: NripeshN <nripesh14@gmail.com>

ivy/functional/frontends/paddle/tensor/tensor.py

@@ -190,6 +190,13 @@ def __iter__(self):
     )
     def __rmul__(self, y, /, name=None):
         return paddle_frontend.multiply(self, y)
+
+    @with_unsupported_dtypes(
+        {"2.5.2 and below": ("bool", "unsigned", "int8", "float16", "bfloat16")},
+        "paddle",
+    )
+    def __float__(self):
+        return float(self._ivy_array)
 
     def __xor__(self, y, /, name=None):
         return paddle_frontend.logic.bitwise_xor(self, y)

ivy_tests/test_ivy/test_frontends/test_paddle/test_tensor/test_tensor.py

@@ -907,6 +907,43 @@ def test_paddle_abs(
     )
 
 
+@handle_frontend_method(
+    class_tree=CLASS_TREE,
+    init_tree="paddle.to_tensor",
+    method_name="__float__",
+    dtype_and_x=helpers.dtype_and_values(
+        available_dtypes=helpers.get_dtypes("numeric"),
+        max_num_dims=0,
+    ),
+)
+def test_paddle__float__(
+    dtype_and_x,
+    frontend_method_data,
+    init_flags,
+    method_flags,
+    backend_fw,
+    frontend,
+    on_device,
+):
+    input_dtypes, xs = dtype_and_x
+    # Numpy doesn't support complex to float conversion
+    assume(not np.issubdtype(input_dtypes[0], np.complexfloating))
+    helpers.test_frontend_method(
+        init_input_dtypes=input_dtypes,
+        backend_to_test=backend_fw,
+        method_input_dtypes=input_dtypes,
+        init_all_as_kwargs_np={
+            "object": xs[0],
+        },
+        method_all_as_kwargs_np={},
+        frontend=frontend,
+        frontend_method_data=frontend_method_data,
+        init_flags=init_flags,
+        method_flags=method_flags,
+        on_device=on_device,
+    )
+
+
 # acosh
 @handle_frontend_method(
     class_tree=CLASS_TREE,