TFLN wafer line is well-positioned for aggressive push into cloud and supercomputing, AI model training and inference, expert says.