Introduce a "Router" layer for providers

Some providers need to be factorized and considered as thin routing
layers on top of other providers. Consider two examples:

- The inference API should be a routing layer over inference providers,
  routed using the "model" key
- The memory banks API is another instance where various memory bank
  types will be provided by independent providers (e.g., a vector store
  is served by Chroma while a keyvalue memory can be served by Redis or
  PGVector)

This commit introduces a generalized routing layer for this purpose.

llama_toolchain/memory/router/__init__.py

@@ -0,0 +1,17 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from typing import Any, List, Tuple
+
+from llama_toolchain.core.datatypes import Api
+
+
+async def get_router_impl(inner_impls: List[Tuple[str, Any]], deps: List[Api]):
+    from .router import MemoryRouterImpl
+
+    impl = MemoryRouterImpl(inner_impls, deps)
+    await impl.initialize()
+    return impl