unregister for memory banks and remove update API (#458)

The semantics of an Update on resources is very tricky to reason about
especially for memory banks and models. The best way to go forward here
is for the user to unregister and register a new resource. We don't have
a compelling reason to support update APIs.


Tests:
pytest -v -s llama_stack/providers/tests/memory/test_memory.py -m
"chroma" --env CHROMA_HOST=localhost --env CHROMA_PORT=8000

pytest -v -s llama_stack/providers/tests/memory/test_memory.py -m
"pgvector" --env PGVECTOR_DB=postgres --env PGVECTOR_USER=postgres --env
PGVECTOR_PASSWORD=mysecretpassword --env PGVECTOR_HOST=0.0.0.0

$CONDA_PREFIX/bin/pytest -v -s -m "ollama"
llama_stack/providers/tests/inference/test_model_registration.py

---------

Co-authored-by: Dinesh Yeduguru <dineshyv@fb.com>

llama_stack/providers/remote/inference/ollama/ollama.py

@@ -93,6 +93,9 @@ class OllamaInferenceAdapter(Inference, ModelsProtocolPrivate):
     async def shutdown(self) -> None:
         pass
 
+    async def unregister_model(self, model_id: str) -> None:
+        pass
+
     async def completion(
         self,
         model_id: str,

llama_stack/providers/remote/inference/tgi/tgi.py

@@ -69,6 +69,9 @@ class _HfAdapter(Inference, ModelsProtocolPrivate):
     async def shutdown(self) -> None:
         pass
 
+    async def unregister_model(self, model_id: str) -> None:
+        pass
+
     async def completion(
         self,
         model: str,

llama_stack/providers/remote/inference/vllm/vllm.py

@@ -58,6 +58,9 @@ class VLLMInferenceAdapter(Inference, ModelsProtocolPrivate):
     async def shutdown(self) -> None:
         pass
 
+    async def unregister_model(self, model_id: str) -> None:
+        pass
+
     async def completion(
         self,
         model_id: str,

llama_stack/providers/remote/memory/chroma/chroma.py

@@ -67,6 +67,9 @@ class ChromaIndex(EmbeddingIndex):
 
         return QueryDocumentsResponse(chunks=chunks, scores=scores)
 
+    async def delete(self):
+        await self.client.delete_collection(self.collection.name)
+
 
 class ChromaMemoryAdapter(Memory, MemoryBanksProtocolPrivate):
     def __init__(self, url: str) -> None:
@@ -134,6 +137,10 @@ class ChromaMemoryAdapter(Memory, MemoryBanksProtocolPrivate):
 
         return [i.bank for i in self.cache.values()]
 
+    async def unregister_memory_bank(self, memory_bank_id: str) -> None:
+        await self.cache[memory_bank_id].index.delete()
+        del self.cache[memory_bank_id]
+
     async def insert_documents(
         self,
         bank_id: str,

llama_stack/providers/remote/memory/pgvector/pgvector.py

@@ -112,6 +112,9 @@ class PGVectorIndex(EmbeddingIndex):
 
         return QueryDocumentsResponse(chunks=chunks, scores=scores)
 
+    async def delete(self):
+        self.cursor.execute(f"DROP TABLE IF EXISTS {self.table_name}")
+
 
 class PGVectorMemoryAdapter(Memory, MemoryBanksProtocolPrivate):
     def __init__(self, config: PGVectorConfig) -> None:
@@ -177,6 +180,10 @@ class PGVectorMemoryAdapter(Memory, MemoryBanksProtocolPrivate):
         )
         self.cache[memory_bank.identifier] = index
 
+    async def unregister_memory_bank(self, memory_bank_id: str) -> None:
+        await self.cache[memory_bank_id].index.delete()
+        del self.cache[memory_bank_id]
+
     async def list_memory_banks(self) -> List[MemoryBank]:
         banks = load_models(self.cursor, VectorMemoryBank)
         for bank in banks: