refactor(coral-tpu): dynamically download deployment models instead of storing in git

Dropping the 13MB of binary .tflite files from the DeepCamera repository
to prevent repository bloat.

The installer scripts (deploy-macos.sh, deploy-linux.sh, and deploy.bat)
now dynamically download the official standard detection models from
Google's Edge TPU GitHub repository at deploy-time using curl (or Start-BitsTransfer
on Windows).

Author: solderzzc

skills/detection/yolo-detection-2026-coral-tpu/.gitignore

@@ -50,3 +50,4 @@ volumes
 
 # ─── Project Specific ────────────────────────────────
 src/yolov7_reid/src/models/mgn_R50-ibn.onnx
+*.tflite

skills/detection/yolo-detection-2026-coral-tpu/deploy-linux.sh

@@ -122,6 +122,14 @@ if ! "$VENV_DIR/bin/python" -m pip install -r "$SKILL_DIR/requirements.txt"; the
     exit 1
 fi
 
+# ─── Download Pre-compiled Models ───────────────────────────────────────────
+emit '{"event": "progress", "stage": "build", "message": "Downloading Edge TPU models..."}'
+mkdir -p "$SKILL_DIR/models"
+cd "$SKILL_DIR/models"
+curl -sSLO "https://github.com/google-coral/edgetpu/raw/master/test_data/ssd_mobilenet_v2_coco_quant_postprocess.tflite"
+curl -sSLO "https://github.com/google-coral/edgetpu/raw/master/test_data/ssd_mobilenet_v2_coco_quant_postprocess_edgetpu.tflite"
+cd "$SKILL_DIR"
+
 log "Linux deployment complete."
 
 # ─── Probe for Edge TPU devices ──────────────────────────────────────────────

skills/detection/yolo-detection-2026-coral-tpu/deploy-macos.sh

@@ -117,6 +117,14 @@ if ! "$VENV_DIR/bin/python" -m pip install --extra-index-url https://google-cora
     exit 1
 fi
 
+# ─── Download Pre-compiled Models ───────────────────────────────────────────
+emit '{"event": "progress", "stage": "build", "message": "Downloading Edge TPU models..."}'
+mkdir -p "$SKILL_DIR/models"
+cd "$SKILL_DIR/models"
+curl -sSLO "https://github.com/google-coral/edgetpu/raw/master/test_data/ssd_mobilenet_v2_coco_quant_postprocess.tflite"
+curl -sSLO "https://github.com/google-coral/edgetpu/raw/master/test_data/ssd_mobilenet_v2_coco_quant_postprocess_edgetpu.tflite"
+cd "$SKILL_DIR"
+
 log "macOS deployment complete."
 
 # ─── Probe for Edge TPU devices ──────────────────────────────────────────────

skills/detection/yolo-detection-2026-coral-tpu/deploy.bat

@@ -116,7 +116,20 @@ if %errorlevel% neq 0 (
 
 echo %LOG_PREFIX% Dependencies installed successfully. 1>&2
 
-REM ─── Step 5: Probe for Edge TPU devices ────────────────────────────────────
+REM ─── Step 5: Download Pre-compiled Models ──────────────────────────────────
+
+echo %LOG_PREFIX% Downloading Edge TPU models... 1>&2
+echo {"event": "progress", "stage": "build", "message": "Downloading Edge TPU models..."}
+
+if not exist "%SKILL_DIR%models" mkdir "%SKILL_DIR%models"
+cd /d "%SKILL_DIR%models"
+
+powershell -Command "Invoke-WebRequest -Uri 'https://github.com/google-coral/edgetpu/raw/master/test_data/ssd_mobilenet_v2_coco_quant_postprocess.tflite' -OutFile 'ssd_mobilenet_v2_coco_quant_postprocess.tflite'"
+powershell -Command "Invoke-WebRequest -Uri 'https://github.com/google-coral/edgetpu/raw/master/test_data/ssd_mobilenet_v2_coco_quant_postprocess_edgetpu.tflite' -OutFile 'ssd_mobilenet_v2_coco_quant_postprocess_edgetpu.tflite'"
+
+cd /d "%SKILL_DIR%"
+
+REM ─── Step 6: Probe for Edge TPU devices ────────────────────────────────────
 
 echo %LOG_PREFIX% Probing for Edge TPU devices natively... 1>&2
 echo {"event": "progress", "stage": "probe", "message": "Checking for physical Edge TPU..."}
@@ -137,7 +150,7 @@ if %errorlevel% equ 0 (
     echo {"event": "progress", "stage": "probe", "message": "No Edge TPU detected -- CPU fallback available"}
 )
 
-REM ─── Step 6: Complete ──────────────────────────────────────────────────────
+REM ─── Step 7: Complete ──────────────────────────────────────────────────────
 
 if "!TPU_FOUND!"=="true" (
     echo {"event": "complete", "status": "success", "tpu_found": true, "message": "Native Coral TPU skill installed -- Edge TPU ready"}