Update on "[Executorch][target recipes] Add target based recipes for lowering models to a target device"

This diff introduces multi backend/ target based recipes to lower a model with very less code. Target recipes provide pre-configured backend recipes to use them and retarget if needed.

See RFC: #13732

## Usage

```
from executorch.export import export
from executorch.export.target_recipes import get_ios_recipe

# CoreML + XNNPACK (FP32)
recipe = get_ios_recipe() # default = "ios-arm64-coreml-fp16"
session = export(model, recipe, example_inputs)
session.save_pte_file("model.pte")
```

## Advanced usage to combine specific recipes of one or two or backends
one can directly use `ExportRecipe.combine_recipes()` to combine specific backend recipes.
```
recipe1 = ExportRecipe.get_recipe(CoreMLRecipeType.FP32)
recipe2 = ExportRecipe.get_recipe(XNNPackRecipeType.FP32)
combined_recipe = ExportRecipe.combine(
                     [recipe1, recipe2],
                     recipe_name="multi_backend_coreml_xnnpack_fp32"
                  )
session = export(model, combined_recipe, example_inputs)
```
Additional changes:
1. Relaxed kwarg validation in the backend providers to just ignore them instead of erroring out.

Fixes: #13732

Differential Revision: [D81297451](https://our.internmc.facebook.com/intern/diff/D81297451/)

[ghstack-poisoned]

Author: abhinaykukkadapu

export/tests/TARGETS

@@ -41,5 +41,7 @@ runtime.python_test(
         "//executorch/export:lib",
         "//executorch/export:target_recipes",
         "//executorch/runtime:runtime",
+        "//executorch/backends/xnnpack/recipes:xnnpack_recipes",
+        "//executorch/backends/apple/coreml:coreml_recipes",
     ]
 )

export/tests/test_target_recipes.py

@@ -10,14 +10,28 @@
 import unittest
 
 import torch
-from executorch.export import export
+from executorch.backends.apple.coreml.recipes import CoreMLRecipeProvider  # pyre-ignore
+from executorch.backends.xnnpack.recipes.xnnpack_recipe_provider import (
+    XNNPACKRecipeProvider,
+)
+from executorch.export import export, recipe_registry
 from executorch.export.target_recipes import get_ios_recipe
 from executorch.runtime import Runtime
 
 
 class TestTargetRecipes(unittest.TestCase):
     """Test target recipes."""
 
+    def setUp(self) -> None:
+        torch._dynamo.reset()
+        super().setUp()
+        recipe_registry.register_backend_recipe_provider(XNNPACKRecipeProvider())
+        # pyre-ignore
+        recipe_registry.register_backend_recipe_provider(CoreMLRecipeProvider())
+
+    def tearDown(self) -> None:
+        super().tearDown()
+
     def test_ios_fp32_recipe_with_xnnpack_fallback(self) -> None:
         # Linear ops skipped by coreml but handled by xnnpack
         class Model(torch.nn.Module):
@@ -93,7 +107,7 @@ def forward(self, x, y):
             et_output = session.run_method("forward", example_inputs[0])
             logging.info(f"et output {et_output}")
 
-    def test_ios_int8_recipe(self) -> None:
+    def test_ios_quant_recipes(self) -> None:
         class Model(torch.nn.Module):
             def __init__(self):
                 super().__init__()