Enable sane naming of registered objects with defaults (#429)

# What does this PR do? 

This is a follow-up to #425. That PR allows for specifying models in the
registry, but each entry needs to look like:

```yaml
- identifier: ...
  provider_id: ...
  provider_resource_identifier: ...
```

This is headache-inducing.

The current PR makes this situation better by adopting the shape of our
APIs. Namely, we need the user to only specify `model-id`. The rest
should be optional and figured out by the Stack. You can always override
it.

Here's what example `ollama` "full stack" registry looks like (we still
need to kill or simplify shield_type crap):
```yaml
models:
- model_id: Llama3.2-3B-Instruct
- model_id: Llama-Guard-3-1B
shields:
- shield_id: llama_guard
  shield_type: llama_guard
```

## Test Plan

See test plan for #425. Re-ran it.

llama_stack/providers/tests/agents/fixtures.py

@@ -9,7 +9,7 @@ import tempfile
 import pytest
 import pytest_asyncio
 
-from llama_stack.apis.models import Model
+from llama_stack.apis.models import ModelInput
 from llama_stack.distribution.datatypes import Api, Provider
 
 from llama_stack.providers.inline.agents.meta_reference import (
@@ -71,13 +71,9 @@ async def agents_stack(request, inference_model, safety_model):
         if fixture.provider_data:
             provider_data.update(fixture.provider_data)
 
-    inf_provider_id = providers["inference"][0].provider_id
-    safety_provider_id = providers["safety"][0].provider_id
-
-    shield = get_shield_to_register(
-        providers["safety"][0].provider_type, safety_provider_id, safety_model
+    shield_input = get_shield_to_register(
+        providers["safety"][0].provider_type, safety_model
     )
-
     inference_models = (
         inference_model if isinstance(inference_model, list) else [inference_model]
     )
@@ -86,13 +82,11 @@ async def agents_stack(request, inference_model, safety_model):
         providers,
         provider_data,
         models=[
-            Model(
-                identifier=model,
-                provider_id=inf_provider_id,
-                provider_resource_id=model,
+            ModelInput(
+                model_id=model,
             )
             for model in inference_models
         ],
-        shields=[shield],
+        shields=[shield_input],
     )
     return impls[Api.agents], impls[Api.memory]

llama_stack/providers/tests/inference/fixtures.py

@@ -9,7 +9,7 @@ import os
 import pytest
 import pytest_asyncio
 
-from llama_stack.apis.models import Model
+from llama_stack.apis.models import ModelInput
 
 from llama_stack.distribution.datatypes import Api, Provider
 from llama_stack.providers.inline.inference.meta_reference import (
@@ -162,10 +162,8 @@ async def inference_stack(request, inference_model):
         {"inference": inference_fixture.providers},
         inference_fixture.provider_data,
         models=[
-            Model(
-                identifier=inference_model,
-                provider_resource_id=inference_model,
-                provider_id=inference_fixture.providers[0].provider_id,
+            ModelInput(
+                model_id=inference_model,
             )
         ],
     )

llama_stack/providers/tests/safety/fixtures.py

@@ -7,9 +7,9 @@
 import pytest
 import pytest_asyncio
 
-from llama_stack.apis.models import Model
+from llama_stack.apis.models import ModelInput
 
-from llama_stack.apis.shields import Shield, ShieldType
+from llama_stack.apis.shields import ShieldInput, ShieldType
 
 from llama_stack.distribution.datatypes import Api, Provider
 from llama_stack.providers.inline.safety.llama_guard import LlamaGuardConfig
@@ -99,28 +99,21 @@ async def safety_stack(inference_model, safety_model, request):
         provider_data.update(safety_fixture.provider_data)
 
     shield_provider_type = safety_fixture.providers[0].provider_type
-    shield = get_shield_to_register(
-        shield_provider_type, safety_fixture.providers[0].provider_id, safety_model
-    )
+    shield_input = get_shield_to_register(shield_provider_type, safety_model)
 
     impls = await resolve_impls_for_test_v2(
         [Api.safety, Api.shields, Api.inference],
         providers,
         provider_data,
-        models=[
-            Model(
-                identifier=inference_model,
-                provider_id=inference_fixture.providers[0].provider_id,
-                provider_resource_id=inference_model,
-            )
-        ],
-        shields=[shield],
+        models=[ModelInput(model_id=inference_model)],
+        shields=[shield_input],
     )
 
+    shield = await impls[Api.shields].get_shield(shield_input.shield_id)
     return impls[Api.safety], impls[Api.shields], shield
 
 
-def get_shield_to_register(provider_type: str, provider_id: str, safety_model: str):
+def get_shield_to_register(provider_type: str, safety_model: str) -> ShieldInput:
     shield_config = {}
     shield_type = ShieldType.llama_guard
     identifier = "llama_guard"
@@ -133,10 +126,8 @@ def get_shield_to_register(provider_type: str, provider_id: str, safety_model: s
         shield_config["guardrailVersion"] = get_env_or_fail("BEDROCK_GUARDRAIL_VERSION")
         shield_type = ShieldType.generic_content_shield
 
-    return Shield(
-        identifier=identifier,
+    return ShieldInput(
+        shield_id=identifier,
         shield_type=shield_type,
         params=shield_config,
-        provider_id=provider_id,
-        provider_resource_id=identifier,
     )

llama_stack/providers/tests/scoring/fixtures.py

@@ -7,6 +7,8 @@
 import pytest
 import pytest_asyncio
 
+from llama_stack.apis.models import ModelInput
+
 from llama_stack.distribution.datatypes import Api, Provider
 
 from llama_stack.providers.tests.resolver import resolve_impls_for_test_v2
@@ -76,20 +78,14 @@ async def scoring_stack(request, inference_model):
         [Api.scoring, Api.datasetio, Api.inference],
         providers,
         provider_data,
-    )
-
-    provider_id = providers["inference"][0].provider_id
-    await impls[Api.models].register_model(
-        model_id=inference_model,
-        provider_id=provider_id,
-    )
-    await impls[Api.models].register_model(
-        model_id="Llama3.1-405B-Instruct",
-        provider_id=provider_id,
-    )
-    await impls[Api.models].register_model(
-        model_id="Llama3.1-8B-Instruct",
-        provider_id=provider_id,
+        models=[
+            ModelInput(model_id=model)
+            for model in [
+                inference_model,
+                "Llama3.1-405B-Instruct",
+                "Llama3.1-8B-Instruct",
+            ]
+        ],
     )
 
     return impls