Closed
Description
Since long we automatically profile the extension host when it becomes unresponsive. In that cases we send a signal via telemetry and in bad cases we ask users to file issues. During that we ask them to attach the corresponding profile but we have mixed results with that. So, in addition to the existing flow we want to log a richer telemetry event that uploads the heaviest sample trace from the profiling