feat: per-phase timings of the frame path in the log

One log line per phase every PHASE_LOG_EVERY samples (default 100, 0 = off):
mean and max of wait_frames, align, convert, each JPEG/PNG encoding,
colormap, depth_stats, base64, json and ws_send. Meant for the measurement
on PC2 via `docker logs` (optimize-streams R01): where a capture spends its
~130 ms of CPU is not known yet.

capture() moves its numpy conversion into _frame_from() unchanged, only so
the block can be timed; _depth_colormap() likewise wraps _jet_colormap().

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

src/backend/config.py

@@ -26,6 +26,10 @@ class Settings(BaseSettings):
     camera_fps: int = 30
     camera_auto_open: bool = True
 
+    # Frame-path timings: one log line per phase every N samples (0 = off).
+    # Read from `docker logs` to see where a capture spends its time on PC2.
+    phase_log_every: int = 100
+
     # YOLO detector (optional — disabled if model file or onnxruntime missing)
     yolo_onnx_path: str = "/app/models/best.onnx"
     yolo_classes: str = ""