feat(agent): add voice input and photo attachments to assistant

Wire mic through Whisper-compatible transcriptions on LLM_BASE_URL.

Photos upload to MinIO and reach the vision model as base64 image_url parts.

src/modules/agent/assistant-chat-storage.ts

@@ -1,9 +1,12 @@
+import type { ClientChatMessage } from "./messages";
+
 export type AssistantChatMessage = {
   role: "user" | "assistant";
   content: string;
+  imageUrl?: string;
 };
 
-const STORAGE_VERSION = "v1";
+const STORAGE_VERSION = "v2";
 const MAX_MESSAGES = 40;
 
 function storageKey(userId: string) {
@@ -13,11 +16,22 @@ function storageKey(userId: string) {
 function isValidMessage(value: unknown): value is AssistantChatMessage {
   if (!value || typeof value !== "object") return false;
   const row = value as Record<string, unknown>;
-  return (
-    (row.role === "user" || row.role === "assistant") &&
-    typeof row.content === "string" &&
-    row.content.trim().length > 0
-  );
+  if (row.role !== "user" && row.role !== "assistant") return false;
+  if (typeof row.content !== "string" || row.content.trim().length === 0) return false;
+  if (row.imageUrl !== undefined && typeof row.imageUrl !== "string") return false;
+  return true;
+}
+
+export function toClientChatMessage(message: AssistantChatMessage): ClientChatMessage {
+  if (!message.imageUrl) {
+    return { role: message.role, content: message.content };
+  }
+
+  return {
+    role: message.role,
+    content: message.content,
+    attachments: [{ type: "image", url: message.imageUrl }],
+  };
 }
 
 export function loadAssistantChat(userId: string): AssistantChatMessage[] {