improve intel gpu stats error handling and remove stale zero-usage warning

Verify an i915/xe device exists in /sys/class/drm before scanning /proc, and log a distinct warning for each durable failure: an intel_gpu_device that does not resolve, one that resolves to a non-Intel card, an unreadable /proc, and a kernel that publishes no per-client engine counters. A present GPU with no attached DRM clients now reports an idle 0% reading instead of latching the hour-long hwaccel error cooldown, which blanked stats after momentary gaps such as camera restarts. Remove the System page warning that blamed intel_gpu_top for all-zero readings, since 0.18 no longer uses intel_gpu_top and 0% is now a legitimate idle reading.

web/public/locales/en/views/system.json

@@ -107,12 +107,7 @@
       },
       "npuUsage": "NPU Usage",
       "npuMemory": "NPU Memory",
-      "npuTemperature": "NPU Temperature",
-      "intelGpuWarning": {
-        "title": "Intel GPU Stats Warning",
-        "message": "GPU stats unavailable",
-        "description": "This is a known bug in Intel's GPU stats reporting tools (intel_gpu_top) where it will break and repeatedly return a GPU usage of 0% even in cases where hardware acceleration and object detection are correctly running on the (i)GPU. This is not a Frigate bug. You can restart the host to temporarily fix the issue and confirm that the GPU is working correctly. This does not affect performance."
-      }
+      "npuTemperature": "NPU Temperature"
     },
     "otherProcesses": {
       "title": "Other Processes",