enhancement: evaluate language detection packages to see which is best for detecting short text

[Coniferish]

Stemming from conversations here and here, it would be worthwhile to do our own comparison of language detection packages to see which is best for detecting the language of short text. Speed and size of the packages should also be considered. Packages of interest include langdetect (which we are currently using), fasttext (if it is compatible with py 3.11), langid, and lingua. We are also currently using a regex pattern and arbitrary text length limit to default to "eng", so this should also be considered/reconsidered.

See detect_languages in lang.py

[Coniferish]

Notes/research from 10/2023

Package	Url	Stars	Last updated	License	Python compatibility	Detects multiple languages in the same text	Deterministic?
Langdetect	https://github.com/Mimino666/langdetect	1.5k	2021
Polyglot	https://github.com/aboSamoor/polyglot	2.2k	2020	GPLv3
~~python-polyglot~~	https://github.com/lainq/polyglot
cld3	https://github.com/google/cld3	706	2022	Apache 2.0
pycld3	https://github.com/bsolomon1124/pycld3	135	2021	Apache 2.0
pycld2	https://github.com/aboSamoor/pycld2	147	2022	Apache 2.0
Fasttext	https://github.com/facebookresearch/fastText	25.1k	2023	MIT
~~spaCy~~	https://github.com/explosion/spaCy
lingua	https://github.com/wichert/lingua	44	2022
lingua-py	https://github.com/pemistahl/lingua-py	635	2023	apache 2.0
~~googletrans~~	https://github.com/ssut/py-googletrans
~~textblob~~	https://github.com/sloria/TextBlob	8.7k	2023
langid	https://github.com/saffsd/langid.py	2.2k	2017	BSD-2-Clause
py3langid	https://github.com/adbar/py3langid	26	2022	BSD 3-Clause	Python >= 3.6		not documented

[jbne]

Thanks for that table. Seems like Polyglot is actually GPLv3, not sure if that was changed recently.