Improve handling of character limits in html stripping The code now more closely keeps track of character counts during html stripping, and should be absolutely exact. When the limit is exceeded, it now restarts the stripping without any URLs to prevent incorrect trimming. It also better preserves whitespace in the original post. New tests are added for Twitter silo to ensure it works as expected.

syntax: glob
*.pyc
*.egg-info
.eggs
.pytest_cache
__pycache__
build
dist
silorider/version.py