chore: remove nested imports

* Since our API packages use import * in __init__.py, we can import
  directly from llama_stack.apis.models instead of
  llama_stack.apis.models.models.  However, the choice to use import *
  is debatable and may need to be reconsidered in the future.

* Remove the unnecessary Ruff F401 suppression.

* Consolidate the Ruff F403 rule configuration in
pyproject.toml.

Signed-off-by: Sébastien Han <seb@redhat.com>

llama_stack/providers/utils/inference/litellm_openai_mixin.py

@@ -23,6 +23,13 @@ from llama_stack.apis.inference import (
     JsonSchemaResponseFormat,
     LogProbConfig,
     Message,
+    OpenAIChatCompletion,
+    OpenAIChatCompletionChunk,
+    OpenAICompletion,
+    OpenAIEmbeddingsResponse,
+    OpenAIEmbeddingUsage,
+    OpenAIMessageParam,
+    OpenAIResponseFormatParam,
     ResponseFormat,
     SamplingParams,
     TextTruncation,
@@ -31,16 +38,7 @@ from llama_stack.apis.inference import (
     ToolDefinition,
     ToolPromptFormat,
 )
-from llama_stack.apis.inference.inference import (
-    OpenAIChatCompletion,
-    OpenAIChatCompletionChunk,
-    OpenAICompletion,
-    OpenAIEmbeddingsResponse,
-    OpenAIEmbeddingUsage,
-    OpenAIMessageParam,
-    OpenAIResponseFormatParam,
-)
-from llama_stack.apis.models.models import Model
+from llama_stack.apis.models import Model
 from llama_stack.distribution.request_headers import NeedsRequestProviderData
 from llama_stack.log import get_logger
 from llama_stack.providers.utils.inference.model_registry import ModelRegistryHelper

llama_stack/providers/utils/inference/model_registry.py

@@ -8,7 +8,7 @@ from typing import Any
 
 from pydantic import BaseModel, Field
 
-from llama_stack.apis.models.models import ModelType
+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
 from llama_stack.providers.utils.inference import (

llama_stack/providers/utils/inference/openai_compat.py

@@ -95,27 +95,25 @@ from llama_stack.apis.inference import (
     CompletionResponse,
     CompletionResponseStreamChunk,
     GreedySamplingStrategy,
-    Message,
-    SamplingParams,
-    SystemMessage,
-    TokenLogProbs,
-    ToolChoice,
-    ToolResponseMessage,
-    TopKSamplingStrategy,
-    TopPSamplingStrategy,
-    UserMessage,
-)
-from llama_stack.apis.inference.inference import (
     JsonSchemaResponseFormat,
+    Message,
     OpenAIChatCompletion,
     OpenAICompletion,
     OpenAICompletionChoice,
     OpenAIEmbeddingData,
     OpenAIMessageParam,
     OpenAIResponseFormatParam,
+    SamplingParams,
+    SystemMessage,
+    TokenLogProbs,
+    ToolChoice,
     ToolConfig,
+    ToolResponseMessage,
+    TopKSamplingStrategy,
+    TopPSamplingStrategy,
+    UserMessage,
 )
-from llama_stack.apis.inference.inference import (
+from llama_stack.apis.inference import (
     OpenAIChoice as OpenAIChatCompletionChoice,
 )
 from llama_stack.models.llama.datatypes import (