remove numeral translation_data

[noviluni]

The numeral_translation_data has been never used. Keeping these files makes the code maintainability harder and it's decreasing the coverage.

I think that these files were created and kept with the idea of being able to parse natural written numbers (for example: "twenty-five" for 25), however, this part will be supplied by a separate library called number-parser: https://github.com/scrapinghub/number-parser. PR: #711

[codecov]

Codecov Report

Merging #782 into master will increase coverage by 2.76%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #782      +/-   ##
==========================================
+ Coverage   95.57%   98.33%   +2.76%     
==========================================
  Files         304      231      -73     
  Lines        2600     2527      -73     
==========================================
  Hits         2485     2485              
+ Misses        115       42      -73     

Impacted Files	Coverage Δ
dateparser/data/__init__.py	100.00% <100.00%> (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 a7ce6c9...a4babd7. Read the comment docs.

[noviluni]

Notice that the total coverage is increased by a tremendous +2.76%. 💪

[Gallaecio]

Nice, 100% is getting closer 🎉