studio: agent window, a chat with a model that edits the scene

✦ Agent in the Studio's top bar opens a floating chat window. The model
(any OpenAI-compatible endpoint with tool calling; default B4L's inference
server, https://inference.p0.contact/v1) gets the 15 MCP tools plus two page
tools (select_object, set_view) and the Studio's state (selection, level,
rotation) in its system prompt. Answers and reasoning stream in; tool calls
show as rows with arguments and results; object ids link to the object;
Stop ends a turn; Undo takes the agent's edits back. The window moves by
its title bar and resizes at its top-left corner.

- studio/agent.js: shared by both front ends (waits for studio.js's
  isokit:studio event); MCP over /mcp or isokit.mcp in the web Studio
- isokit studio: /api/agent/{config,models,chat} relay the calls on their
  own threads (ureq, chunked and flushed per read), so the key stays on the
  machine and the endpoint needs no CORS; settings from ISOKIT_LLM_URL /
  _KEY / _MODEL or agent.json in the user's config folder
- web Studio: calls the endpoint from the page with a key kept in the
  browser; build_web.py --agent-url names a same-site relay as default; the
  deploy builds with /llm/v1, which deploy/Caddyfile forwards to the
  inference server (its nginx refuses the CORS preflight)
- tools/mock_llm.py + tools/agent_checks.py: the agent window checks in both
  check_studio.py and check_web.py

docs/ROADMAP.md

@@ -36,6 +36,19 @@ Each phase ends with tests, a Studio/viewer check in Edge, commit and push.
 - [x] `tools/build_web.py` (static output), `tools/check_web.py` (44 checks, Edge)
 - [ ] Deploy a hosted copy; render in a Web Worker for very large scenes
 
+## Agent window (done)
+
+- [x] ✦ Agent in the Studio: chat with a model on an OpenAI-compatible endpoint (default: B4L's
+      inference server) that edits the scene with the MCP tools + page tools (select, view);
+      streamed answers and reasoning, tool calls with results, clickable object ids, Stop, Undo
+- [x] `isokit studio` relays the calls (`/api/agent/*`, key stays on the machine, no CORS);
+      settings from `ISOKIT_LLM_*` or `agent.json` in the user's config folder
+- [x] Web Studio calls the endpoint from the page; isokit.buildfor.life relays `/llm/` to the
+      inference server (Caddy) so the page needs no CORS from it
+- [x] `tools/mock_llm.py` + `tools/agent_checks.py` in both Studio checks
+- [ ] Keep the conversation per project across reloads; attach a rendered view (PNG, Phase 4)
+      for vision models
+
 ## Phase 2: richer editing
 
 - [ ] Interior prefabs (desk, table with chairs, shelf, machine, bed, sofa, partition walls) and a