Add shared ai-service container as AI provider intermediary

All feature containers now POST messages to ai-service (port 8010) instead
of calling AI providers directly. ai-service routes to LM Studio, Ollama,
or Anthropic based on /config/ai_service_config.json. doc-service AI
providers removed; replaced by httpx ai_client.py. Backend settings
restructured to /api/settings/ai. Frontend gets dedicated AIAdminSettingsPage
and AI Service card in AppsPage.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

frontend/src/api/client.ts

@@ -168,18 +168,21 @@ export interface AIProviderUpdate {
   lmstudio_api_key?: string;
 }
 
-export const getDocumentSettings = () =>
-  api.get<Record<string, unknown>>("/settings/documents").then((r) => r.data);
+export const getAISettings = () =>
+  api.get<Record<string, unknown>>("/settings/ai").then((r) => r.data);
 
-export const updateDocumentAISettings = (data: AIProviderUpdate) =>
-  api.patch<Record<string, unknown>>("/settings/documents/ai", data).then((r) => r.data);
+export const updateAISettings = (data: AIProviderUpdate) =>
+  api.patch<Record<string, unknown>>("/settings/ai", data).then((r) => r.data);
 
-export const testDocumentAIConnection = () =>
+export const testAIConnection = () =>
   api.post<{ ok: boolean; provider: string; response?: string; error?: string }>(
-    "/settings/documents/ai/test"
+    "/settings/ai/test"
   ).then((r) => r.data);
 
 export const updateDocumentLimits = (max_pdf_mb: number) =>
   api.patch<Record<string, unknown>>("/settings/documents/limits", { max_pdf_mb }).then(
     (r) => r.data
   );
+
+export const getDocumentLimits = () =>
+  api.get<Record<string, unknown>>("/settings/documents/limits").then((r) => r.data);