[pt] Dictionary update to v1.0.1

[p-goulart]

In this version...

v1.0.0

• 🇬🇧 check English words in the speller (this is pretty major);
• 📧 email is finally accepted alongside e-mail, with hyphenation enforced as a picky rule (closes "email" with no hyphen is also correct portuguese-pos-dict#27)
• ✨ add new words from Please add words #10503 as well as new words from our Grafana dashboard from the last 60 days;

v1.0.1 (bugfix)

• 😩 fix an annoying bug that meant synthesis of 'mesmo' also led to 'mesmíssimo' forms generated (affected constructions are not detected by gGEC).

English spelling 🇬🇧 🇺🇸

We use the following heuristics:

• we use a word list to decide if a specific token is likely to be part of a longer string of English words;
• we tag adjacent words as English words...
  • if they are also in the word list, or
  • if they are untagged (i.e. probably absent from the Portuguese dictionary) and they exist in English (we use the English tagger to determine this).

If the English tagger dictionary lookup fails, the word is not valid in English, and we do not tag it as _english_ignore_. This means that, later, we run the Portuguese speller on such words.

The original version of my idea went even further. I wanted to tag all such words and, later, actually get suggestions for them from the English speller.

Currently, if a user misspells a word in English, they will be probably irrelevant suggestions in Portuguese, rather than the more relevant ones in English. This could be a future improvement.

---

^{Lol ignore the branch name. Originally this was going to be v0.16, but @maphjo pointed out it was probably time to revisit the versioning scheme. This wouldn't normally be a major release, as it introduces no breaking changes, but conceptually it makes sense to start with v1.0.0 finally, and we're not planning any breaking changes to the dictionary binaries in the near future.}