Update on "[ExecuTorch][#10375] Add extension.BundledModule to Wrap extension.Module with Bundled Program Logic"

#10375

# Context

This issue is a step of #9638.

In #9638, we want to have `extension.Module` as the single source of implementation in `pybindings`, which means that `pybindings.PyModule` should use `extension.Module` rather than its own `pybindings.Module`.

The issue is that `pybindings.PyModule` is dependent on the `method` getter from `pybindings.Module`, which `extension.Module` do not have. Since we don't want to expose `method` getter in `extension.Module`, we have to protect the getter, wrap the functions that is dependent on it and use the protected getter there, ultimately decouple `pybindings` from a `method` getter.

# Proposal

Now that we have a protected `method` getter, we can introduce a `extension.BundledModule`, a child class inheriting `extension.Module` which wraps up bundled program logic that is dependent on the `method` getter.

Differential Revision: [D73564125](https://our.internmc.facebook.com/intern/diff/D73564125/)

[ghstack-poisoned]

Author: zhenyan-zhang-meta

extension/module/test/resources/gen_bundled_program.py

@@ -10,6 +10,7 @@
 from executorch.exir import to_edge_transform_and_lower
 from torch.export import export, export_for_training
 
+
 # Step 1: ExecuTorch Program Export
 class SampleModel(torch.nn.Module):
     """An example model with multi-methods. Each method has multiple input and single output"""
@@ -27,6 +28,7 @@ def forward(self, x: torch.Tensor, q: torch.Tensor) -> torch.Tensor:
         torch.add(y, q, out=y)
         return y
 
+
 def main() -> None:
     """Sample code to generate bundled program and save it to file. It is the same as in https://pytorch.org/executorch/0.6/bundled-io.html#emit-example"""
     # Inference method name of SampleModel we want to bundle testcases to.
@@ -46,7 +48,6 @@ def main() -> None:
         capture_input,
     )
 
-
     # Emit the traced method into ET Program.
     et_program = to_edge_transform_and_lower(method_graph).to_executorch()