refactor(metrics): MetricsProvider abstraction with Android /proc backend

Moves direct psutil.* calls behind a MetricsProvider Protocol so the
codebase no longer needs ad-hoc `if psutil is not None` guards at every
call site. Each provider lives in its own module under
utils/metrics/: PsutilMetricsProvider for desktop, NullMetricsProvider
as a zeroed fallback, AndroidMetricsProvider that reads /proc/stat,
/proc/meminfo, /proc/self/stat, and /proc/self/status directly (psutil
isn't available under Chaquopy). The Android provider tracks the
previous CPU sample so cpu_percent() returns delta-based percentages
matching psutil's interval=None semantics, and degrades to zeros when
any /proc file is unreadable instead of crashing the dashboard.

Factory get_metrics_provider() in utils/metrics/__init__.py picks
Android > psutil > Null. api/routes/system.py and
core/processing/metrics_history.py now go through the factory; psutil
import is confined to one place. 12 new unit tests cover paren-in-comm
parsing of /proc/self/stat, delta CPU%, missing-file resilience, and
factory selection order. Full suite: 727 passing.

TODO.md

@@ -68,22 +68,15 @@ Drive USB LED controllers (APA102, WS2812) connected directly to the Android TV
 
 ## Performance Metrics Abstraction
 
-The codebase has direct `psutil.*` calls scattered across `api/routes/system.py` and `core/processing/metrics_history.py`, with ad-hoc `if psutil is not None` guards sprinkled in to support Android. This couples Android platform handling to every call site.
+- [x] `MetricsProvider` protocol + dataclass DTOs (`MemorySnapshot`, `ProcessSnapshot`) live in `server/src/ledgrab/utils/metrics/types.py`. Each provider has its own module: `psutil_provider.py`, `null_provider.py`, `android_provider.py`.
+- [x] Factory `get_metrics_provider()` in `utils/metrics/__init__.py` selects Android → psutil → Null. `psutil` import is now confined to one place.
+- [x] `api/routes/system.py` and `core/processing/metrics_history.py` use the provider; no more `if psutil is not None` guards in the hot paths.
+- [x] Android `/proc`-backed provider implemented (`/proc/stat`, `/proc/meminfo`, `/proc/self/stat`, `/proc/self/status`). Carries previous-sample state for delta-based CPU%; degrades to zeros if any `/proc` file is locked down. 12 unit tests cover both desktop and Android paths.
 
-- [ ] Refactor: introduce `MetricsProvider` protocol in `utils/metrics.py` with methods like `cpu_percent()`, `memory_info()`, `process_info()`
-- [ ] Implement `PsutilMetricsProvider` (desktop) and `NullMetricsProvider` (fallback when psutil missing)
-- [ ] Later: `AndroidMetricsProvider` reading from `/proc` (see section below)
-- [ ] Replace all direct `psutil.*` calls with the provider; only one factory location knows about psutil availability
+## Android Performance Metrics — Future Enhancements
 
-## Android Performance Metrics
+Beyond the `/proc`-based AndroidMetricsProvider that's now in place:
 
-Currently `psutil` (used for CPU/RAM monitoring in the web UI) is not available on Android via Chaquopy. Metrics calls are guarded with `if psutil is not None` so they return no data on Android.
-
-- [ ] Implement Android-native metrics collection:
-  - CPU usage via `/proc/self/stat` + `/proc/stat` parsing (no psutil needed)
-  - RAM usage via `/proc/meminfo` or `ActivityManager.getMemoryInfo()` through Chaquopy bridge
-  - App-specific memory via `Debug.getMemoryInfo()` (Kotlin → Python)
-- [ ] Create `AndroidMetricsProvider` in `server/src/ledgrab/utils/` that implements the same interface as the psutil-based provider
-- [ ] Wire into existing metrics endpoints (`/api/v1/system/metrics`) with platform detection
+- [ ] Optional: app-specific memory via `Debug.getMemoryInfo()` through a Kotlin → Python Chaquopy bridge (more accurate than `VmRSS` for split-app-process accounting)
 - [ ] Consider: device battery/temperature readings for TV boxes (some have thermal throttling)
 - [ ] Optional: GPU usage via `/sys/class/kgsl/kgsl-3d0/gpubusy` on Adreno, Mali-specific paths for Mali GPUs