Introduce missing APIs to lower ExportedProgram objects directly

Summary: Currently, there is an imbalance in compilation APIs where the `export_to_executorch` function is used on a module, although that module is already from an exported program. This diff introduces APIs to further lower ExportedPrograms directly, without going back to module.

Differential Revision: D70529392

Author: mcremon-meta

backends/cadence/aot/compiler.py

@@ -178,18 +178,11 @@ def export_program(
     return expo_program
 
 
-# Export the model and lower it to an EdgeProgramManager (in edge IR).
-def export_to_edge(
-    model: torch.nn.Module,
-    inputs: tuple[object, ...],
+def lower_ep_to_edge(
+    expo_program: ExportedProgram,
     dump_graphs: bool = False,
     constant_methods: Optional[dict[str, object]] = None,
 ) -> EdgeProgramManager:
-    assert isinstance(model, torch.nn.Module), "model should be an nn.Module"
-
-    # Export the model into an ExportedProgram.
-    expo_program = export_program(model, inputs)
-
     # Call to_edge to convert the graph to edge IR.
     # Note: dim_order is skipped (https://github.com/pytorch/executorch/issues/3704)
     edge_prog_manager = to_edge(
@@ -213,6 +206,23 @@ def export_to_edge(
         logging.info(
             edge_prog_manager.exported_program().graph_module.graph.print_tabular()
         )
+    return edge_prog_manager
+
+
+# Export the model and lower it to an EdgeProgramManager (in edge IR).
+def export_to_edge(
+    model: torch.nn.Module,
+    inputs: tuple[object, ...],
+    dump_graphs: bool = False,
+    constant_methods: Optional[dict[str, object]] = None,
+) -> EdgeProgramManager:
+    assert isinstance(model, torch.nn.Module), "model should be an nn.Module"
+
+    # Export the model into an ExportedProgram.
+    expo_program = export_program(model, inputs)
+
+    # Lower the model to edge IR.
+    edge_prog_manager = lower_ep_to_edge(expo_program, dump_graphs, constant_methods)
 
     return edge_prog_manager