Update how we input kwargs

Summary:
Previously, the code for passing inputs to exported program was:
```
if kwargs:
    return (args, kwargs)
else:
    return args
```

However, this causes some inconsistency where if the original input contains args and kwargs, the treespec would be a tuple containing a tuple of arguments, and a dictionary of keyword arguments. But if the original input only contained args, the treespec would just be a tuple of arguments.

So I updated the code to just always keep the kwargs around

cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx chenyang78 aakhundov kadeng

X-link: pytorch/pytorch#109160

Differential Revision: D49218534

Pulled By: angelayi

Author: angelayi

exir/capture/_capture.py

@@ -163,7 +163,7 @@ def capture(  # noqa: C901
         # input spec, but due to some limitations in pytree implementation, it doesn't
         # recognize the make_fx graph with torchdynamo input spec. We workaround it
         # by getting the input spec directly from user argument.
-        in_spec = pytree.tree_flatten(args)[1]
+        in_spec = pytree.tree_flatten((args, {}))[1]
 
         if config.enable_functionalization and not config.enable_aot:
             args = copy.deepcopy(args)