Add type hints to torch/_functorch files pt 2 (#173460)

Summary:
Add comprehensive type annotations to 2 files in torch/_functorch that were previously missing type coverage. Each file now passes pyrefly check with stricter sub-config settings (implicit-any=true).

Files updated:

aot_autograd.py
eager_transforms.py

Authored with help of Claude Code

X-link: pytorch/pytorch#173460
Approved by: https://github.com/aorenste

Reviewed By: wdvr

Differential Revision: D91862651

fbshipit-source-id: 13483cf6d59428c84e41c9d37ba8e0ed17305e5b

Author: Lucaskabela

userbenchmark/dynamo/dynamobench/_dynamo/utils.py

@@ -3853,7 +3853,7 @@ def get_real_value(node: torch.fx.Node, tracer: Any) -> Any:
     return real_value
 
 
-def assert_no_fake_params_or_buffers(gm: torch.fx.GraphModule) -> None:
+def assert_no_fake_params_or_buffers(gm: torch.nn.Module) -> None:
     from torch._subclasses.fake_tensor import FakeTensorConfig, is_fake
 
     def stack_or_hint(t: Any) -> str: