Combining accents

[mnater]

Playing around with Int.Segmenter (#145) reveleaded an issue with combining accents.
E.g. the word "БЕЛАРУ́СКАЯ" in test6.html contains the letter "У" with a "COMBINING ACUTE ACCENT ◌́").

The current implementation for finding words first looks for consecutive characters of the unicode property escape \p{Letter} and then checks if this set contains a character that is not in the alphabet defined by the .wasm file of the respective language.
Since combining accents are not part of \p{Letter} and there is no normalized character the current implementation finds "БЕЛАР" and hyphenates just this part of the word. This could lead to errors.

How to solve (ideas):

• Include \p{Mn} (or a subset) in the regex at line 562 (-> don't hyphenate this word at all)
• Also include \p{Mn} in the "alphabet" but omit while hyphenating