Python: Upgrade Minimum Onnx Version to enable MacOS Unit Tests (#9981)

Closes : #9979

enabled test cases for mac os
upgraded uv.lock file

### Motivation and Context

Using version 0.4.0 did not provide a pip package for MacOS, which
forced us to disabled Unit Tests on MacOs
With version 0.5.0 available we can enable the unit Tests for MacOS.

Using Version 0.5.0 will enable following features for users :

1. Phi3.5 and Phi3.5 MoE
2. MacOS Support without Building the Code from Source
3. LoRa Adapter Swapping

### Contribution Checklist

<!-- Before submitting this PR, please make sure: -->

- [x] The code builds clean without any errors or warnings
- [x] The PR follows the [SK Contribution
Guidelines](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md)
and the [pre-submission formatting
script](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md#development-scripts)
raises no violations
- [x] All unit tests pass, and I have added new tests where possible
- [x] I didn't break anyone 😄

---------

Co-authored-by: Eduard van Valkenburg <[email protected]>

python/pyproject.toml

@@ -86,7 +86,7 @@ ollama = [
     "ollama ~= 0.4"
 ]
 onnx = [
-    "onnxruntime-genai ~= 0.4; platform_system != 'Darwin'"
+    "onnxruntime-genai ~= 0.5"
 ]
 anthropic = [
     "anthropic ~= 0.32"
@@ -156,7 +156,8 @@ filterwarnings = [
 ]
 timeout = 120
 markers = [
-    "ollama: mark a test as requiring the Ollama service (use \"not ollama\" to skip those tests)"
+    "ollama: mark a test as requiring the Ollama service (use \"not ollama\" to skip those tests)",
+    "onnx: mark a test as requiring the Onnx service (use \"not onnx\" to skip those tests)"
 ]
 
 [tool.ruff]

python/tests/integration/completions/chat_completion_test_base.py

@@ -2,7 +2,6 @@
 
 
 import os
-import platform
 import sys
 from typing import Annotated
 
@@ -22,6 +21,7 @@
 from semantic_kernel.connectors.ai.google.vertex_ai import VertexAIChatCompletion, VertexAIChatPromptExecutionSettings
 from semantic_kernel.connectors.ai.mistral_ai import MistralAIChatCompletion, MistralAIChatPromptExecutionSettings
 from semantic_kernel.connectors.ai.ollama import OllamaChatCompletion, OllamaChatPromptExecutionSettings
+from semantic_kernel.connectors.ai.onnx import OnnxGenAIChatCompletion, OnnxGenAIPromptExecutionSettings, ONNXTemplate
 from semantic_kernel.connectors.ai.open_ai import (
     AzureChatCompletion,
     AzureChatPromptExecutionSettings,
@@ -71,12 +71,6 @@
 bedrock_setup: bool = is_service_setup_for_testing(["AWS_DEFAULT_REGION"], raise_if_not_set=False)
 
 
-skip_on_mac_available = platform.system() == "Darwin"
-if not skip_on_mac_available:
-    from semantic_kernel.connectors.ai.onnx import OnnxGenAIChatCompletion, OnnxGenAIPromptExecutionSettings
-    from semantic_kernel.connectors.ai.onnx.utils import ONNXTemplate
-
-
 # A mock plugin that contains a function that returns a complex object.
 class PersonDetails(KernelBaseModel):
     id: str
@@ -155,7 +149,7 @@ def services(self) -> dict[str, tuple[ServiceType | None, type[PromptExecutionSe
             "vertex_ai": (VertexAIChatCompletion() if vertex_ai_setup else None, VertexAIChatPromptExecutionSettings),
             "onnx_gen_ai": (
                 OnnxGenAIChatCompletion(template=ONNXTemplate.PHI3V) if onnx_setup else None,
-                OnnxGenAIPromptExecutionSettings if not skip_on_mac_available else None,
+                OnnxGenAIPromptExecutionSettings,
             ),
             "bedrock_amazon_titan": (
                 BedrockChatCompletion(model_id="amazon.titan-text-premier-v1:0") if bedrock_setup else None,

python/tests/integration/completions/test_chat_completion_with_image_input_text_output.py

@@ -118,7 +118,10 @@
                 ChatMessageContent(role=AuthorRole.USER, items=[TextContent(text="Where was it made?")]),
             ],
             {},
-            marks=pytest.mark.skipif(not onnx_setup, reason="Need a Onnx Model setup"),
+            marks=(
+                pytest.mark.skipif(not onnx_setup, reason="Need a Onnx Model setup"),
+                pytest.mark.onnx,
+            ),
             id="onnx_gen_ai_image_input_file",
         ),
         pytest.param(

python/tests/integration/completions/test_chat_completions.py

@@ -150,7 +150,10 @@ class Reasoning(KernelBaseModel):
                 ChatMessageContent(role=AuthorRole.USER, items=[TextContent(text="How are you today?")]),
             ],
             {},
-            marks=pytest.mark.skipif(not onnx_setup, reason="Need a Onnx Model setup"),
+            marks=(
+                pytest.mark.skipif(not onnx_setup, reason="Need a Onnx Model setup"),
+                pytest.mark.onnx,
+            ),
             id="onnx_gen_ai",
         ),
         # endregion

python/tests/integration/completions/test_text_completion.py

@@ -1,6 +1,5 @@
 # Copyright (c) Microsoft. All rights reserved.
 
-import platform
 import sys
 from functools import partial
 from typing import Any
@@ -19,6 +18,7 @@
 from semantic_kernel.connectors.ai.google.vertex_ai import VertexAITextCompletion, VertexAITextPromptExecutionSettings
 from semantic_kernel.connectors.ai.hugging_face import HuggingFacePromptExecutionSettings, HuggingFaceTextCompletion
 from semantic_kernel.connectors.ai.ollama import OllamaTextCompletion, OllamaTextPromptExecutionSettings
+from semantic_kernel.connectors.ai.onnx import OnnxGenAIPromptExecutionSettings, OnnxGenAITextCompletion
 from semantic_kernel.connectors.ai.open_ai import (
     AzureOpenAISettings,
     AzureTextCompletion,
@@ -43,11 +43,6 @@
 )  # Tests are optional for ONNX
 bedrock_setup = is_service_setup_for_testing(["AWS_DEFAULT_REGION"], raise_if_not_set=False)
 
-skip_on_mac_available = platform.system() == "Darwin"
-if not skip_on_mac_available:
-    from semantic_kernel.connectors.ai.onnx import OnnxGenAIPromptExecutionSettings, OnnxGenAITextCompletion
-
-
 pytestmark = pytest.mark.parametrize(
     "service_id, execution_settings_kwargs, inputs, kwargs",
     [
@@ -128,7 +123,10 @@
             {},
             ["<|user|>Repeat the word Hello<|end|><|assistant|>"],
             {},
-            marks=pytest.mark.skipif(not onnx_setup, reason="Need local Onnx setup"),
+            marks=(
+                pytest.mark.skipif(not onnx_setup, reason="Need a Onnx Model setup"),
+                pytest.mark.onnx,
+            ),
             id="onnx_gen_ai_text_completion",
         ),
         pytest.param(
@@ -242,7 +240,7 @@ def services(self) -> dict[str, tuple[ServiceType | None, type[PromptExecutionSe
             ),
             "onnx_gen_ai": (
                 OnnxGenAITextCompletion() if onnx_setup else None,
-                OnnxGenAIPromptExecutionSettings if not skip_on_mac_available else None,
+                OnnxGenAIPromptExecutionSettings,
             ),
             # Amazon Bedrock supports models from multiple providers but requests to and responses from the models are
             # inconsistent. So we need to test each model separately.