Add support for multiple ZIM langs + validate Language metadata

[benoit74]

Fix #300

Changes:

• isolate method to parse language received either from input (CLI) args or automatically retrieved from HTML metadata
• add unit testing on this method
• add support for a list of languages, permissive to spaces which will be trimmed and preserving list of languages order
• add validation of language metadata, including early validation if the lang comes from input (CLI) args
• add a real fallback to eng as indicated in usage, wasn't done when no lang was provided at CLI and no lang was found in WARC records

[codecov[bot]]

Codecov Report

Attention: Patch coverage is 92.85714% with 2 lines in your changes missing coverage. Please review.

Project coverage is 84.17%. Comparing base (44e3f8c) to head (86c69f6).

Files	Patch %	Lines
src/warc2zim/converter.py	81.81%	2 Missing :warning:

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #304 +/- ## ========================================== + Coverage 83.94% 84.17% +0.23% ========================================== Files 13 14 +1 Lines 1227 1245 +18 Branches 235 240 +5 ========================================== + Hits 1030 1048 +18 Misses 152 152 Partials 45 45 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[benoit74]

I've also changed log level to WARNING since these problems are recovered, and I added a WARNING log when the general fallback is applied (when no language has been specified at input - CLI - and no language has been found in main entry)