Fix qualcomm passes (#6110)

Summary:
Pull Request resolved: #6110

A new pass is added after #5920

bypass-github-export-checks

Reviewed By: tarun292

Differential Revision: D64190781

fbshipit-source-id: a2fe9dedf502f7ff743a9f37ea8bee5f190227ea

Author: cccclai

…/passes/expand_broadcast_tensor_shape.py …_passes/expand_broadcast_tensor_shape.pybackends/qualcomm/passes/expand_broadcast_tensor_shape.py renamed to backends/qualcomm/_passes/expand_broadcast_tensor_shape.py backends/qualcomm/passes/expand_broadcast_tensor_shape.py renamed to backends/qualcomm/_passes/expand_broadcast_tensor_shape.py

@@ -30,6 +30,9 @@
 )
 from executorch.backends.qualcomm._passes.convert_prelu import ConvertPReLU
 from executorch.backends.qualcomm._passes.convert_to_linear import ConvertToLinear
+from executorch.backends.qualcomm._passes.expand_broadcast_tensor_shape import (
+    ExpandBroadcastTensorShape,
+)
 from executorch.backends.qualcomm._passes.fold_qdq import FoldQDQ
 from executorch.backends.qualcomm._passes.i64_to_i32 import I64toI32
 from executorch.backends.qualcomm._passes.layout_transform import LayoutTransform
@@ -47,9 +50,6 @@
     QNN_TENSOR_TYPE_MAP,
 )
 from executorch.backends.qualcomm.builders.qnn_constants import OpContextLoader
-from executorch.backends.qualcomm.passes.expand_broadcast_tensor_shape import (
-    ExpandBroadcastTensorShape,
-)
 from executorch.backends.qualcomm.serialization.qnn_compile_spec_schema import (
     _soc_info_table,
     QcomChipset,