Clarify Web Tasks Vision telemetry and Pose Landmarker model redistribution terms

revelado auto

`@mediapipe/tasks-genai@0.10.27` — `gemma-4-E4B-it-web.task` loads but decode output is numeric tokens, not text

[LLM][Android 16] SIGSEGV in libllm_inference_engine_jni.so on first generate() after model load (Pixel 6 Pro, tasks-genai 0.10.35)

Document PyInstaller/macos packaging for MediaPipe Python Tasks apps

SIGSEGV in AndroidPacketCreator.nativeCreateRgbaImage called from ModelResourcesCacheService.nativeInstallServiceObject

Google AI Edge Gallery - Gemma3-1B-IT

`RuntimeError: memory access out of bounds` loading gemma-4-E2B-it-web.task on Chrome 146 / macOS / Apple M4

`RCE via unsafe deserialization in Model Maker (torch.load, Keras, tar slip, yaml.FullLoader)`

🐛 Issue: Inconsistent Hand Landmark Detection / Glitch in Live Stream Mode (Android)