[Unity] Enhance Torch-consistency in rehsape (#16360)

This PR introduces the following signature changes:

- `Tensor.reshape(shape)` to `Tensor.reshape(*shape)`
- `Tensor.permute_dims(axes)` to `Tensor.permute_dims(*axes)`

Author: junrushao

python/tvm/relax/frontend/nn/_tensor_op.py

@@ -78,10 +78,10 @@ def minimum(self, other):
         other = _convert_scalar(other, self)
         return _op().minimum(self, other)
 
-    def reshape(self, shape):
+    def reshape(self, *shape):
         return _op().reshape(self, shape)
 
-    def permute_dims(self, axes):
+    def permute_dims(self, *axes):
         return _op().permute_dims(self, axes)
 
     def repeat(self, repeats: int, axis: Optional[int] = None):

tests/python/relax/test_frontend_nn_tensor.py

@@ -14,15 +14,15 @@
 # KIND, either express or implied.  See the License for the
 # specific language governing permissions and limitations
 # under the License.
+import numpy as np
 import pytest
+
 import tvm
 import tvm.testing
 from tvm import relax
-from tvm.relax.frontend.nn import Tensor, Module, spec
+from tvm.relax.frontend.nn import Module, Tensor, spec
 from tvm.script import relax as R
 
-import numpy as np
-
 
 def test_tensor_from_numpy():
     x = np.random.rand(1, 10)
@@ -136,8 +136,8 @@ def test(x: R.Tensor((1, 10), dtype="float32"), _io: R.Object) -> R.Tuple(R.Tens
 def test_tensor_op_manipulate():
     class Model(Module):
         def test(self, x: Tensor):
-            z0 = x.reshape([2, 5, 2])
-            z1 = x.permute_dims([2, 1, 0])
+            z0 = x.reshape(2, 5, 2)
+            z1 = x.permute_dims(2, 1, 0)
             z2 = x.repeat(2, axis=1)
             return (z0, z1, z2)