refactor: enforce top-level imports for llama-stack-api

Enforce that all imports from llama-stack-api use the form:

from llama_stack_api import <symbol>

 This prevents external code from accessing internal package structure
 (e.g., llama_stack_api.agents, llama_stack_api.common.*) and establishes
 a clear public API boundary.

 Changes:
 - Export 400+ symbols from llama_stack_api/__init__.py
 - Include all API types, common utilities, and strong_typing helpers
 - Update files across src/llama_stack, docs/, tests/, scripts/
 - Convert all submodule imports to top-level imports
 - ensure docs use the proper importing structure

 Addresses PR review feedback requiring explicit __all__ definition to
 prevent "peeking inside" the API package.

Signed-off-by: Charlie Doern <cdoern@redhat.com>

src/llama_stack/core/stack.py

@@ -12,27 +12,30 @@ import tempfile
 from typing import Any
 
 import yaml
-from llama_stack_api.agents import Agents
-from llama_stack_api.batches import Batches
-from llama_stack_api.benchmarks import Benchmarks
-from llama_stack_api.conversations import Conversations
-from llama_stack_api.datasetio import DatasetIO
-from llama_stack_api.datasets import Datasets
-from llama_stack_api.datatypes import Api
-from llama_stack_api.eval import Eval
-from llama_stack_api.files import Files
-from llama_stack_api.inference import Inference
-from llama_stack_api.inspect import Inspect
-from llama_stack_api.models import Models
-from llama_stack_api.post_training import PostTraining
-from llama_stack_api.prompts import Prompts
-from llama_stack_api.providers import Providers
-from llama_stack_api.safety import Safety
-from llama_stack_api.scoring import Scoring
-from llama_stack_api.scoring_functions import ScoringFunctions
-from llama_stack_api.shields import Shields
-from llama_stack_api.tools import ToolGroups, ToolRuntime
-from llama_stack_api.vector_io import VectorIO
+from llama_stack_api import (
+    Agents,
+    Api,
+    Batches,
+    Benchmarks,
+    Conversations,
+    DatasetIO,
+    Datasets,
+    Eval,
+    Files,
+    Inference,
+    Inspect,
+    Models,
+    PostTraining,
+    Prompts,
+    Providers,
+    Safety,
+    Scoring,
+    ScoringFunctions,
+    Shields,
+    ToolGroups,
+    ToolRuntime,
+    VectorIO,
+)
 
 from llama_stack.core.conversations.conversations import ConversationServiceConfig, ConversationServiceImpl
 from llama_stack.core.datatypes import Provider, SafetyConfig, StackRunConfig, VectorStoresConfig