feat(analytics): instance census, full capture, richer error context (#511)

Add a once-per-boot instance_started event (arch, os, deploy_mode,
gpu_present) so the fleet architecture mix is measurable. It reuses the
existing per-instance instance_id and is exempt from the volume sample
rate, since a census that fires once per boot must not be thinned.

Restore useful capture depth now that the sponsored plan removes the
quota pressure behind the earlier hardening:

- PostHog sample rate 0.1 to 1.0 (full analytics when enabled); the
  property allowlist still blocks file data.
- Sentry per-instance ceiling 20 to 500/hr, breadcrumb trail restored
  (sanitized: urls/paths redacted, data payloads dropped), full stack
  paths kept; local vars, request bodies, and PII still dropped. Both
  api and web.

Honor ANALYTICS_ENABLED=false as an opt-out alias: it was documented on
the Docker Hub README but never wired in 2.x, so anyone who set it was
still tracked.

All capture stays behind the analytics opt-out gate.

tests/unit/api/system-info.test.ts

@@ -0,0 +1,57 @@
+import { afterEach, describe, expect, it, vi } from "vitest";
+
+const mockExistsSync = vi.hoisted(() => vi.fn());
+const mockDeployMode = vi.hoisted(() => vi.fn());
+
+vi.mock("node:fs", () => ({
+  existsSync: mockExistsSync,
+}));
+vi.mock("../../../apps/api/src/lib/deploy-mode.js", () => ({
+  deployMode: mockDeployMode,
+}));
+
+import { gatherSystemProperties } from "../../../apps/api/src/lib/system-info.js";
+
+describe("gatherSystemProperties", () => {
+  const origArch = process.arch;
+
+  afterEach(() => {
+    Object.defineProperty(process, "arch", { value: origArch, configurable: true });
+    mockExistsSync.mockReset();
+    mockDeployMode.mockReset();
+  });
+
+  it("reports arm64 for an arm64 process", () => {
+    Object.defineProperty(process, "arch", { value: "arm64", configurable: true });
+    mockDeployMode.mockReturnValue("native");
+    mockExistsSync.mockReturnValue(false);
+    expect(gatherSystemProperties().arch).toBe("arm64");
+  });
+
+  it("reports amd64 for an x64 process", () => {
+    Object.defineProperty(process, "arch", { value: "x64", configurable: true });
+    mockDeployMode.mockReturnValue("native");
+    mockExistsSync.mockReturnValue(false);
+    expect(gatherSystemProperties().arch).toBe("amd64");
+  });
+
+  it("reports gpu_present true when /dev/nvidia0 exists", () => {
+    mockExistsSync.mockImplementation((p: string) => p === "/dev/nvidia0");
+    mockDeployMode.mockReturnValue("external");
+    expect(gatherSystemProperties().gpu_present).toBe(true);
+  });
+
+  it("reports gpu_present false when /dev/nvidia0 is absent", () => {
+    mockExistsSync.mockReturnValue(false);
+    mockDeployMode.mockReturnValue("external");
+    expect(gatherSystemProperties().gpu_present).toBe(false);
+  });
+
+  it("passes through deploy mode and os platform", () => {
+    mockDeployMode.mockReturnValue("embedded");
+    mockExistsSync.mockReturnValue(false);
+    const props = gatherSystemProperties();
+    expect(props.deploy_mode).toBe("embedded");
+    expect(props.os_platform).toBe(process.platform);
+  });
+});