feat(watchers): add box stability watcher daemon, recovery guardrails, and CLI integration

- Add dedicated watchers/ project folder with box-stability-watcher.py supervisor
- Monitor host load, memory, swap saturation, and crash-looping services
- Implement tiered mitigations: yellow renicing, orange SIGSTOP pause with 60s grace, red shedding
- Distinguish user-launched agents (allowed on desktop default socket) from automated box workloads
- Wire first-class box stability CLI subcommand and top-line host status in fleet status
- Harden tmux.service with cgroup memory limits to prevent OS freeze and OOM avalanches
- Add 10-test unit test suite covering thresholds, safety whitelist, pause/resume, and isolation

watchers/box-stability.json

@@ -0,0 +1,44 @@
+{
+  "thresholds": {
+    "load_warning": 20.0,
+    "load_critical": 35.0,
+    "load_emergency": 60.0,
+    "ram_warning_pct": 80.0,
+    "ram_critical_pct": 90.0,
+    "swap_warning_pct": 75.0,
+    "swap_critical_pct": 85.0,
+    "process_rss_warning_mb": 2000,
+    "process_rss_critical_mb": 3000
+  },
+  "protected_commands": [
+    "sshd",
+    "tailscaled",
+    "tailscale",
+    "systemd",
+    "dbus-broker",
+    "pipewire",
+    "wireplumber",
+    "tmux",
+    "bash",
+    "zsh",
+    "ghostty",
+    "alacritty"
+  ],
+  "monitored_targets": [
+    "muse-bin",
+    "chromium",
+    "python3",
+    "parec"
+  ],
+  "actions": {
+    "enable_renice": true,
+    "enable_cull_runaway": true,
+    "enable_service_freeze": true
+  },
+  "service_guardrails": {
+    "max_restarts_per_window": 10,
+    "window_seconds": 60
+  },
+  "interval_sec": 10,
+  "log_file": "logs/box-stability.jsonl"
+}