chore: standardize unsupported model error #2517 (#2518)

# What does this PR do?

- llama_stack/exceptions.py: Add UnsupportedModelError class
- remote inference ollama.py and utils/inference/model_registry.py:
Changed ValueError in favor of UnsupportedModelError
- utils/inference/litellm_openai_mixin.py: remove `register_model`
function implementation from `LiteLLMOpenAIMixin` class. Now uses the
parent class `ModelRegistryHelper`'s function implementation

Closes #2517


## Test Plan


1. Create a new `test_run_openai.yaml` and paste the following config in
it:

```yaml
version: '2'
image_name: test-image
apis:
- inference
providers:
  inference:
  - provider_id: openai
    provider_type: remote::openai
    config:
      max_tokens: 8192
models:
- metadata: {}
  model_id: "non-existent-model"
  provider_id: openai
  model_type: llm
server:
  port: 8321
```

And run the server with:
```bash
uv run llama stack run test_run_openai.yaml
```

You should now get a `llama_stack.exceptions.UnsupportedModelError` with
the supported list of models in the error message.

---

Tested for the following remote inference providers, and they all raise
the `UnsupportedModelError`:
- Anthropic
- Cerebras
- Fireworks
- Gemini
- Groq
- Ollama
- OpenAI
- SambaNova
- Together
- Watsonx

---------

Co-authored-by: Rohan Awhad <rawhad@redhat.com>

llama_stack/apis/common/errors.py

@@ -0,0 +1,13 @@
+# 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.
+
+
+class UnsupportedModelError(ValueError):
+    """raised when model is not present in the list of supported models"""
+
+    def __init__(self, model_name: str, supported_models_list: list[str]):
+        message = f"'{model_name}' model is not supported. Supported models are: {', '.join(supported_models_list)}"
+        super().__init__(message)

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

@@ -18,6 +18,7 @@ from llama_stack.apis.common.content_types import (
     InterleavedContentItem,
     TextContentItem,
 )
+from llama_stack.apis.common.errors import UnsupportedModelError
 from llama_stack.apis.inference import (
     ChatCompletionRequest,
     ChatCompletionResponse,
@@ -376,9 +377,7 @@ class OllamaInferenceAdapter(
                     f"Imprecise provider resource id was used but 'latest' is available in Ollama - using '{model.provider_resource_id}:latest'"
                 )
                 return model
-            raise ValueError(
+            raise UnsupportedModelError(model.provider_resource_id, available_models)
-                f"Model '{model.provider_resource_id}' is not available in Ollama. Available models: {', '.join(available_models)}"
-            )
         model.provider_resource_id = provider_resource_id
 
         return model

llama_stack/providers/utils/inference/litellm_openai_mixin.py

@@ -13,6 +13,7 @@ from llama_stack.apis.common.content_types import (
     InterleavedContent,
     InterleavedContentItem,
 )
+from llama_stack.apis.common.errors import UnsupportedModelError
 from llama_stack.apis.inference import (
     ChatCompletionRequest,
     ChatCompletionResponse,
@@ -92,7 +93,7 @@ class LiteLLMOpenAIMixin(
     async def register_model(self, model: Model) -> Model:
         model_id = self.get_provider_model_id(model.provider_resource_id)
         if model_id is None:
-            raise ValueError(f"Unsupported model: {model.provider_resource_id}")
+            raise UnsupportedModelError(model.provider_resource_id, self.alias_to_provider_id_map.keys())
         return model
 
     def get_litellm_model_name(self, model_id: str) -> str:

llama_stack/providers/utils/inference/model_registry.py

@@ -8,6 +8,7 @@ from typing import Any
 
 from pydantic import BaseModel, Field
 
+from llama_stack.apis.common.errors import UnsupportedModelError
 from llama_stack.apis.models import ModelType
 from llama_stack.models.llama.sku_list import all_registered_models
 from llama_stack.providers.datatypes import Model, ModelsProtocolPrivate
@@ -34,7 +35,9 @@ def get_huggingface_repo(model_descriptor: str) -> str | None:
 
 
 def build_hf_repo_model_entry(
-    provider_model_id: str, model_descriptor: str, additional_aliases: list[str] | None = None
+    provider_model_id: str,
+    model_descriptor: str,
+    additional_aliases: list[str] | None = None,
 ) -> ProviderModelEntry:
     aliases = [
         get_huggingface_repo(model_descriptor),
@@ -81,9 +84,7 @@ class ModelRegistryHelper(ModelsProtocolPrivate):
 
     async def register_model(self, model: Model) -> Model:
         if not (supported_model_id := self.get_provider_model_id(model.provider_resource_id)):
-            raise ValueError(
+            raise UnsupportedModelError(model.provider_resource_id, self.alias_to_provider_id_map.keys())
-                f"Model '{model.provider_resource_id}' is not supported. Supported models are: {', '.join(self.alias_to_provider_id_map.keys())}"
-            )
         provider_resource_id = self.get_provider_model_id(model.model_id)
         if model.model_type == ModelType.embedding:
             # embedding models are always registered by their provider model id and does not need to be mapped to a llama model