fix(resources): Create bo_punct_position.csv

[ngawangtrinley]

• Opening punctuations are appended at the left end of sentences (i.e. ༈→བཀྲ་ཤིས་ཤོག)
• Closing punctuations are appended at the right end of sentences (i.e. བཀྲ་ཤིས་ཤོག←།).
• Spaces are always grouped with punctuation characters (i.e. ༈→བཀྲ་ཤིས་ཤོག)
• Groups of punctuation characters containing one or more opening punctuation are treated as opening punctuation. (i.e. ༄༅། །→བཀྲ་ཤིས་ཤོག)
• Groups of punctuation characters NOT containing one or more opening punctuation are treated as closing punctuation. (i.e. བཀྲ་ཤིས་ཤོག←། །།)

KNOWN LIMITATION: there are exceptions to the above punctuation group rules such as ། ༈ ། which contains both opening (༈) and closing (།) punctuation characters but needs to be handled by split into ། ། and ༈, respectively a closing and an opening puntuation group. However these cases are not very common and are ignored for now.

[eroux]

just for reference, I have some code from a few years ago that handles punctuation: https://github.com/buda-base/git-to-dbs/blob/master/src/main/java/io/bdrc/gittodbs/TibetanStringChunker.java