Fix parse issue with consecutive dates

[kiskoza]

Closes #58

The issue was with the offset used for extra_chars_to_be_added. The inserted string is 12 characters long, but the code incremented the offset by only 10 - it should have been 11 (we're replacing 1 char with 12 chars, 12 - 1 = 11).

Why wasn't it a problem? The output variable holds the json with extra spaces added, so it had some space to "correct" itself.

# `output` argument, note the double spaces
'{"first_date":  "2016-01-25",  "second_date":  "2014-01-26",  "third_date":  "2024-01-02"}'
# `output` after the first insert - looks okay
'{"first_date": !!timestamp "2016-01-25",  "second_date":  "2014-01-26",  "third_date":  "2024-01-02"}'
# `output` after the second insert - a bit off, but still valid
'{"first_date": !!timestamp "2016-01-25",  "second_date":!!timestamp  "2014-01-26",  "third_date":  "2024-01-02"}'
# `output` after the third insert - the extra keyword ate the `:` and the json is not valid anymore
'{"first_date": !!timestamp "2016-01-25",  "second_date":!!timestamp  "2014-01-26",  "third_date"!!timestamp   "2024-01-02"}'

There's a green pipeline using my other PR: https://github.com/kiskoza/crack/actions/runs/7386253493