Set capture_autograd_function=False by default (pytorch#141932)

pytorch#136959 cleaned up the flag and added a warning. @Chillee pointed out that we should really default this flag to false otherwise we subject all users that go down this code path to log spew.

Pull Request resolved: pytorch#141932
Approved by: https://github.com/jansel

Author: bobrenjc93

torch/_dynamo/config.py

@@ -428,7 +428,7 @@ def default_debug_dir_root():
 only_allow_pt2_compliant_ops = False
 
 # This flag is ignored and maintained for backwards compatibility.
-capture_autograd_function = True
+capture_autograd_function = False
 
 # This flag is ignored and maintained for backwards compatbility.
 capture_func_transforms = True