Fix SimpleFSDP aot_eager_autobucket (#1963)

Stack from [ghstack](https://github.com/ezyang/ghstack/tree/0.12.0)
(oldest at bottom):
* #1960
* #1959
* __->__ #1963

pytorch/pytorch#166130 changes the configs and
this PR adopts the new configs

Squash and Merge button won't work for this PR. I'll merge by myself.

Author: fegin

torchtitan/experiments/simple_fsdp/backend.py

@@ -21,12 +21,14 @@ def get_compile_backend(backend_name: str) -> Union[str, callable]:
         # Perform auto optimization in aten fx-level and execute code in aot_eager backend
         # The autobucketing logic is here: https://github.com/pytorch/pytorch/pull/163960
         from torch._dynamo.backends.common import aot_autograd as aot_autograd_backend
+
+        from torch._inductor.config import aten_distributed_optimizations as dist_opts
         from torch._inductor.fx_passes.overlap_scheduling import (
             schedule_overlap_bucketing,
         )
 
-        torch._inductor.config.test_configs.aten_fx_overlap_preserving_bucketing = True
-        torch._inductor.config.test_configs.aten_fx_overlap_insert_overlap_deps = False
+        dist_opts.collective_bucketing = True
+        dist_opts.insert_overlap_deps = False
         torch._inductor.config.allow_buffer_reuse = False
 
         def aten_autobucketing_reordering_pass(