feat: add static embedding metadata to dynamic model listings for providers using OpenAIMixin (#3547)

# What does this PR do?

- remove auto-download of ollama embedding models
- add embedding model metadata to dynamic listing w/ unit test
- add support and tests for allowed_models
- removed inference provider models.py files where dynamic listing is
enabled
- store embedding metadata in embedding_model_metadata field on
inference providers
- make model_entries optional on ModelRegistryHelper and
LiteLLMOpenAIMixin
- make OpenAIMixin a ModelRegistryHelper
- skip base64 embedding test for remote::ollama, always returns floats
- only use OpenAI client for ollama model listing
- remove unused build_model_entry function
- remove unused get_huggingface_repo function


## Test Plan

ci w/ new tests

llama_stack/providers/remote/inference/vertexai/models.py

@@ -1,20 +0,0 @@
-# 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.
-
-from llama_stack.providers.utils.inference.model_registry import (
-    ProviderModelEntry,
-)
-
-# Vertex AI model IDs with vertex_ai/ prefix as required by litellm
-LLM_MODEL_IDS = [
-    "vertex_ai/gemini-2.0-flash",
-    "vertex_ai/gemini-2.5-flash",
-    "vertex_ai/gemini-2.5-pro",
-]
-
-SAFETY_MODELS_ENTRIES = list[ProviderModelEntry]()
-
-MODEL_ENTRIES = [ProviderModelEntry(provider_model_id=m) for m in LLM_MODEL_IDS] + SAFETY_MODELS_ENTRIES

llama_stack/providers/remote/inference/vertexai/vertexai.py

@@ -16,14 +16,12 @@ from llama_stack.providers.utils.inference.litellm_openai_mixin import (
 from llama_stack.providers.utils.inference.openai_mixin import OpenAIMixin
 
 from .config import VertexAIConfig
-from .models import MODEL_ENTRIES
 
 
 class VertexAIInferenceAdapter(OpenAIMixin, LiteLLMOpenAIMixin):
     def __init__(self, config: VertexAIConfig) -> None:
         LiteLLMOpenAIMixin.__init__(
             self,
-            MODEL_ENTRIES,
             litellm_provider_name="vertex_ai",
             api_key_from_config=None,  # Vertex AI uses ADC, not API keys
             provider_data_api_key_field="vertex_project",  # Use project for validation