Update usage of PyTorch's custom op API

Hi, I maintain the custom ops story in PyTorch. This PR updates the the
usage of PyTorch's private custom op API to a newer API. This API is still
private but closer to what we want it to be.

Test Plan:
- wait for CI

Author: zou3519

py/torch_tensorrt/dynamo/lowering/substitutions/einsum.py

@@ -1,26 +1,21 @@
 from typing import Any, Dict, Optional, Sequence, Tuple
 
 import torch
-from torch._custom_op.impl import custom_op
+import torch._custom_ops as library
 from torch.fx.node import Argument, Target
 from torch_tensorrt.dynamo.lowering._pre_aot_lowering import register_substitution
 from torch_tensorrt.fx.converter_registry import tensorrt_converter
 from torch_tensorrt.fx.converters.converter_utils import set_layer_name
 from torch_tensorrt.fx.types import TRTNetwork, TRTTensor
 
 
-@custom_op(
-    qualname="tensorrt::einsum",
-    manual_schema="(str equation, Tensor[] tensors) -> Tensor",
+library.custom_op(
+    "tensorrt::einsum",
+    "(str equation, Tensor[] tensors) -> Tensor",
 )
-def einsum(equation, tensors):  # type: ignore[no-untyped-def]
-    # Defines operator schema, name, namespace, and function header
-    ...
 
-
-@einsum.impl("cpu")  # type: ignore[misc]
-@einsum.impl("cuda")  # type: ignore[misc]
-@einsum.impl_abstract()  # type: ignore[misc]
+@library.impl("tensorrt::einsum")  # type: ignore[misc]
+@library.impl_abstract("tensorrt::einsum")  # type: ignore[misc]
 def einsum_generic(
     *args: Any,
     **kwargs: Any,

py/torch_tensorrt/dynamo/lowering/substitutions/maxpool1d.py

@@ -1,7 +1,7 @@
 from typing import Any, Dict, Optional, Tuple
 
 import torch
-from torch._custom_op.impl import custom_op
+import torch._custom_ops as library
 from torch.fx.node import Argument, Target
 from torch_tensorrt.dynamo.lowering._pre_aot_lowering import register_substitution
 from torch_tensorrt.fx.converter_registry import tensorrt_converter
@@ -20,14 +20,10 @@
 # types. The namespace, such as tensorrt, will cause the op to be registered as torch.ops.tensorrt.your_op
 # Then, create a placeholder function with no operations, but having the same schema and naming as that
 # used in the decorator
-@custom_op(
-    qualname="tensorrt::maxpool1d",
-    manual_schema="(Tensor x, int[1] kernel_size, int[1] stride, int[1] padding, int[1] dilation, bool ceil_mode) -> Tensor",
+library.custom_op(
+    "tensorrt::maxpool1d",
+    "(Tensor x, int[1] kernel_size, int[1] stride, int[1] padding, int[1] dilation, bool ceil_mode) -> Tensor"
 )
-def maxpool1d(x, kernel_size, stride, padding, dilation, ceil_mode):  # type: ignore[no-untyped-def]
-    # Defines operator schema, name, namespace, and function header
-    ...
-
 
 # 2. The Generic Implementation
 #
@@ -36,9 +32,8 @@ def maxpool1d(x, kernel_size, stride, padding, dilation, ceil_mode):  # type: ig
 # is desirable. If the operator to replace is a custom module you've written, then add its Torch
 # implementation here. Note that the function header to the generic function can have specific arguments
 # as in the above placeholder
-@maxpool1d.impl("cpu")  # type: ignore[misc]
-@maxpool1d.impl("cuda")  # type: ignore[misc]
-@maxpool1d.impl_abstract()  # type: ignore[misc]
+@library.impl("tensorrt::maxpool1d")  # type: ignore[misc]
+@library.impl_abstract("tensorrt::maxpool1d")  # type: ignore[misc]
 def maxpool1d_generic(
     *args: Any,
     **kwargs: Any,