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/lib/llm/openai-compatible.ts

@@ -1,8 +1,8 @@
-import type { ChatCompletionRequest, ChatCompletionResult, LlmClient } from "./types";
+import type { ChatCompletionRequest, ChatCompletionResult, ChatMessage, LlmClient } from "./types";
 
 type OpenAiMessage = {
   role: string;
-  content: string | null;
+  content: ChatMessage["content"];
   tool_calls?: Array<{
     id: string;
     type: "function";