feat: extract pagetype from og:type or ld+json

[andremacola]

This implements #307

• Extract from og:type
• If no og:type tag, try to extract from ld+json
• On some sites the variable element_text was coming with bad html formatting causing error and having to use extract_json_parse_error. This patch fix the problem with html.unescape()
• Tests were done but I'm not sure if I should put validation in all extract_meta_json tests

OBS: I realized that json_metadata already had some @type implementations but I preferred not to change it to prevent break anything and implemented it with new conditions and lists (not sure if this is the ideal approach)

OBS2: I think txttocsv function and test_txttocsv should be refactored. Every minor change in metadata breaks the test.

OBS3: extract_meta_json function could be simplified and refactored

OBS4: Only returns pagetype from ld+json if the page is some kind of an article/category/home/site/bloc etc... got types from https://schema.org/docs/full.html

[codecov-commenter]

Codecov Report

Merging #310 (fed9da9) into master (dd2d212) will decrease coverage by 0.02%.
The diff coverage is 100.00%.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

@@            Coverage Diff             @@
##           master     #310      +/-   ##
==========================================
- Coverage   96.17%   96.15%   -0.02%     
==========================================
  Files          21       21              
  Lines        3241     3253      +12     
==========================================
+ Hits         3117     3128      +11     
- Misses        124      125       +1     

Impacted Files	Coverage Δ
trafilatura/utils.py	97.88% <ø> (-0.53%)	⬇️
trafilatura/json_metadata.py	96.80% <100.00%> (+0.24%)	⬆️
trafilatura/metadata.py	98.43% <100.00%> (+0.01%)	⬆️
trafilatura/core.py	98.10% <0.00%> (ø)

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[adbar]

Thanks for the PR, I'm going to review it. As for your remarks:

1. @type in metadata: @felipehertzer do you have an opinion on this?
2. Yes, txttocsv() could use some refactoring, I'm open for a PR on this btw.
3. I'm not sure how to refactor extract_meta_json(), could you provide a suggestion?
4. Sounds good.