At #2549 @asvetlov proposed useful improvement of detection of content encoding to prevent reading the whole response in memory by allowing to specify max data size for sniffing.

UniversalDetector can be used: https://chardet.readthedocs.io/en/latest/usage.html#example-detecting-encoding-incrementally.