validate DOIs in references

[kmccurley]

At the very minimum, I found that some authors in issue 3 of CiC had used DOIs with the prefix 10.5555. These are "test only" DOIs that should not be used and will not resolve. They may have done it to silence the warnings, or they may have done it because someone gave them 10.5555 DOIs. Apparently ACM is guilty of this. Whenever we parse the bibtex we should check to make sure that they don't start with 10.5555 (just throw them out if they are found).