Some books cannot be imported from KOReader due to the language filter being too strict

[joaquimbc]

Describe the bug
Chinese books are often distribued as a raw .txt file, which means that KOReader didn't set a language for it. Even after altering the language on the metadata file as instructed in the setup guide, no words come up whatsoever and the program gets stuck in a loop. Even more strangely, it says that the vocabulary builder has 0 words saved, which is not true at all. It doesn't even loads the words from the lookup_history file, but it does for the taiwanese epub book, which doesn't makes much sense as they're the same language

To Reproduce
Steps to reproduce the behavior:

1. Load any Chinese .txt file and save something
2. Try to import it

Expected behavior
It should load

Screenshots
If applicable, add screenshots to help explain your problem.

Logs
Go to Help->View session logs, paste the log here. Be sure to redact out any sensitive information. The log can contain clipboard content.

Desktop (please complete the following information):

• OS: Windows
• Vocabsieve version (if nightly, must be latest): 0.12.4

Additional context
I tried changing it to zh, zh-CH, zh-TW and nothing works. Idk if it's just a problem due to the language filter or if the app fails to consider a .txt file altogether, because they don't even show in the logs (the book that shows in the log is a different book, an epub taiwanese one with the zh-TW flag).

metadata.txt
VocabSieve-log.txt

[1over137]

.txt is just not on the file type list so it is not scanned. I guess I can add it thought, there isn't anything wrong with it as long as you have the metadata.