Profile patch

[LinguList]

We have been using code to generate language-specific profiles for some time now. The code that we used to make language-specific profiles form "normal" profiles was always copy-pasted, and it is time to add it to pylexibank.

The code is EXTREMELY easy: we load the profile as a dictionary, we iterate over Graphemes in the data, and we add each IPA segment we find there.

The typical workflow now is:

1. make a general profile for a dataset
2. use cldfbench lexibank.language_profiles DATASET to make language-specific profiles
3. edit those profiles

[LinguList]

I have still not really learned how to do tests in these commandline apps. So excuse me for not adding, but if one of you could add exemplary tests later, I would add them next time.

[codecov-commenter]

Codecov Report

Merging #256 (9a289b0) into master (eeb1c92) will decrease coverage by 1.02%. The diff coverage is 34.28%.

@@            Coverage Diff             @@
##           master     #256      +/-   ##
==========================================
- Coverage   97.19%   96.16%   -1.03%     
==========================================
  Files          32       33       +1     
  Lines        2105     2140      +35     
==========================================
+ Hits         2046     2058      +12     
- Misses         59       82      +23     

Impacted Files	Coverage Δ
src/pylexibank/commands/check_profile.py	50.00% <0.00%> (-1.32%)	:arrow_down:
src/pylexibank/commands/language_profiles.py	36.36% <36.36%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update eeb1c92...9a289b0. Read the comment docs.

[LinguList]

Thanks, these have been addressed now, I throw an error for hillburmish, as we cannot make a profile for individual languages there.