feat(browser): vision automation loop with screenshots, coordinate actions and bounded challenge solving

- browser.screenshot returns the page as an image content block (≤1280px wide, JPEG quality
  configurable) for vision-capable upstreams; non-vision upstreams get a redacted text snapshot.
  Login pages / password fields are marked sensitive and withheld unless a
  browser.sensitive_screenshot grant exists and allowSensitive is passed.
- browser.click_at / type_at / scroll / wait act in screenshot pixels; the server converts via
  viewport/DPR, requires automation control and an allowlisted host, and writes a thumbnail
  evidence file (DATA_DIR/tmp/evidence) referenced from agent_run_event.
- browser.solve_challenge runs a bounded child agent_run (vision caps only, maxSteps ≤ 12,
  budget), judges success from page facts, stops immediately on 2FA / SMS / device confirmation,
  and otherwise moves the session to waiting with evidence and a takeover link.
- PiAgentRuntime passes modelImage outputs as pi-ai image content when model.input has image;
  the ledger never stores image data; coordinate actions use per-invocation idempotency keys.
- Tests: image codec, runtime image content, E14 e2e (login solved, 2FA stop, step limit,
  sensitive withheld). Real camofox 1.17.0 check on example.com recorded in the runbook.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

apps/server/test/e2e/e2-agent-delegate.test.ts

@@ -45,8 +45,8 @@ describe.skipIf(!hasPg)("E2 Agent 委托闭环(fake 模型 + 真实能力/授权
     return r.body.data.id as string;
   };
 
-  it("启动断言:29 个能力全部注册,并在能力目录里可见", async () => {
-    expect(e.booted.ctx.capabilities.list()).toHaveLength(29);
+  it("启动断言:35 个能力全部注册,并在能力目录里可见", async () => {
+    expect(e.booted.ctx.capabilities.list()).toHaveLength(35);
     const r = await e.api("GET", "/capabilities");
     expect(r.body.data.items.map((c: { id: string }) => c.id).sort()).toEqual(
       e.booted.ctx.capabilities