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/groq/groq.py

@@ -9,8 +9,6 @@ from llama_stack.providers.remote.inference.groq.config import GroqConfig
 from llama_stack.providers.utils.inference.litellm_openai_mixin import LiteLLMOpenAIMixin
 from llama_stack.providers.utils.inference.openai_mixin import OpenAIMixin
 
-from .models import MODEL_ENTRIES
-
 
 class GroqInferenceAdapter(OpenAIMixin, LiteLLMOpenAIMixin):
     _config: GroqConfig
@@ -18,7 +16,6 @@ class GroqInferenceAdapter(OpenAIMixin, LiteLLMOpenAIMixin):
     def __init__(self, config: GroqConfig):
         LiteLLMOpenAIMixin.__init__(
             self,
-            model_entries=MODEL_ENTRIES,
             litellm_provider_name="groq",
             api_key_from_config=config.api_key,
             provider_data_api_key_field="groq_api_key",

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

@@ -1,48 +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.models.llama.sku_list import CoreModelId
-from llama_stack.providers.utils.inference.model_registry import (
-    build_hf_repo_model_entry,
-    build_model_entry,
-)
-
-SAFETY_MODELS_ENTRIES = []
-
-MODEL_ENTRIES = [
-    build_hf_repo_model_entry(
-        "llama3-8b-8192",
-        CoreModelId.llama3_1_8b_instruct.value,
-    ),
-    build_model_entry(
-        "llama-3.1-8b-instant",
-        CoreModelId.llama3_1_8b_instruct.value,
-    ),
-    build_hf_repo_model_entry(
-        "llama3-70b-8192",
-        CoreModelId.llama3_70b_instruct.value,
-    ),
-    build_hf_repo_model_entry(
-        "llama-3.3-70b-versatile",
-        CoreModelId.llama3_3_70b_instruct.value,
-    ),
-    # Groq only contains a preview version for llama-3.2-3b
-    # Preview models aren't recommended for production use, but we include this one
-    # to pass the test fixture
-    # TODO(aidand): Replace this with a stable model once Groq supports it
-    build_hf_repo_model_entry(
-        "llama-3.2-3b-preview",
-        CoreModelId.llama3_2_3b_instruct.value,
-    ),
-    build_hf_repo_model_entry(
-        "meta-llama/llama-4-scout-17b-16e-instruct",
-        CoreModelId.llama4_scout_17b_16e_instruct.value,
-    ),
-    build_hf_repo_model_entry(
-        "meta-llama/llama-4-maverick-17b-128e-instruct",
-        CoreModelId.llama4_maverick_17b_128e_instruct.value,
-    ),
-] + SAFETY_MODELS_ENTRIES