Explicit encoding

[dmerejkowsky]

Fix #89

[dmerejkowsky]

I've also added a semgrep rule so that we always specify specific encodings in the future

[cgestes]

The patch seems to override the encoding defined in the system to force utf-8.

Maybe I dont fully understand the situation or the pyrhon side of things but it looks like the bug this patch is supposed to fix is as follow: the system is configured with an encoding different than utf-8, the file being read is utf-8 (or not faulty using an utf8 encoding).

Are all the files processed by tbump utf-8? Or python doing magic when passed encoding= utf-8?
If not I would rather allow overriding the default system encoding, but make it optional.

[cgestes]

Also a path to explore could be to work on raw content.
Is encoding needed?

[dmerejkowsky]

Meh. This was mostly an excuse to try out semgrep. And it works :)

I'm still not sure about the root cause of #89, so let's close this