ER: Use the data written rather than the data read to report progress

[pbatard]

As per pbatard/rufus#580. This may require figuring out the expected extracted size beforehand though...

[pbatard]

This will be difficult to add, as we need to know the uncompressed size of the file, and some of the formats we use (.Z, .bz2) are stream compression formats and do not store that information. This means, we'd have to uncompress the whole thing to get the size.

Only .zip and .gz seem to provide the uncompressed size in an easy to read manner (but even then, there is no guarantee that this data will present for gzip).
As to .lzma (and by extension .xz), you'd have to uncompress part of the file to get that info...

Considering this, the progress information will have to remain read-based.