Update the "InterleavedTextMedia" type (#635)

## What does this PR do?

This is a long-pending change and particularly important to get done
now.

Specifically:
- we cannot "localize" (aka download) any URLs from media attachments
anywhere near our modeling code. it must be done within llama-stack.
- `PIL.Image` is infesting all our APIs via `ImageMedia ->
InterleavedTextMedia` and that cannot be right at all. Anything in the
API surface must be "naturally serializable". We need a standard `{
type: "image", image_url: "<...>" }` which is more extensible
- `UserMessage`, `SystemMessage`, etc. are moved completely to
llama-stack from the llama-models repository.

See https://github.com/meta-llama/llama-models/pull/244 for the
corresponding PR in llama-models.

## Test Plan

```bash
cd llama_stack/providers/tests

pytest -s -v -k "fireworks or ollama or together" inference/test_vision_inference.py
pytest -s -v -k "(fireworks or ollama or together) and llama_3b" inference/test_text_inference.py
pytest -s -v -k chroma memory/test_memory.py \
  --env EMBEDDING_DIMENSION=384 --env CHROMA_DB_PATH=/tmp/foobar

pytest -s -v -k fireworks agents/test_agents.py  \
   --safety-shield=meta-llama/Llama-Guard-3-8B \
   --inference-model=meta-llama/Llama-3.1-8B-Instruct
```

Updated the client sdk (see PR ...), installed the SDK in the same
environment and then ran the SDK tests:

```bash
cd tests/client-sdk
LLAMA_STACK_CONFIG=together pytest -s -v agents/test_agents.py
LLAMA_STACK_CONFIG=ollama pytest -s -v memory/test_memory.py

# this one needed a bit of hacking in the run.yaml to ensure I could register the vision model correctly
INFERENCE_MODEL=llama3.2-vision:latest LLAMA_STACK_CONFIG=ollama pytest -s -v inference/test_inference.py
```

llama_stack/providers/inline/agents/meta_reference/agent_instance.py

@@ -26,6 +26,7 @@ from llama_stack.apis.memory_banks import *  # noqa: F403
 from llama_stack.apis.safety import *  # noqa: F403
 
 from llama_stack.providers.utils.kvstore import KVStore
+from llama_stack.providers.utils.memory.vector_store import concat_interleaved_content
 from llama_stack.providers.utils.telemetry import tracing
 
 from .persistence import AgentPersistence
@@ -389,7 +390,7 @@ class ChatAgent(ShieldRunnerMixin):
 
             if rag_context:
                 last_message = input_messages[-1]
-                last_message.context = "\n".join(rag_context)
+                last_message.context = rag_context
 
         elif attachments and AgentTool.code_interpreter.value in enabled_tools:
             urls = [a.content for a in attachments if isinstance(a.content, URL)]
@@ -655,7 +656,7 @@ class ChatAgent(ShieldRunnerMixin):
 
     async def _retrieve_context(
         self, session_id: str, messages: List[Message], attachments: List[Attachment]
-    ) -> Tuple[Optional[List[str]], Optional[List[int]]]:  # (rag_context, bank_ids)
+    ) -> Tuple[Optional[InterleavedContent], List[int]]:  # (rag_context, bank_ids)
         bank_ids = []
 
         memory = self._memory_tool_definition()
@@ -723,11 +724,16 @@ class ChatAgent(ShieldRunnerMixin):
                 break
             picked.append(f"id:{c.document_id}; content:{c.content}")
 
-        return [
-            "Here are the retrieved documents for relevant context:\n=== START-RETRIEVED-CONTEXT ===\n",
-            *picked,
-            "\n=== END-RETRIEVED-CONTEXT ===\n",
-        ], bank_ids
+        return (
+            concat_interleaved_content(
+                [
+                    "Here are the retrieved documents for relevant context:\n=== START-RETRIEVED-CONTEXT ===\n",
+                    *picked,
+                    "\n=== END-RETRIEVED-CONTEXT ===\n",
+                ]
+            ),
+            bank_ids,
+        )
 
     def _get_tools(self) -> List[ToolDefinition]:
         ret = []