delete stig now deletes records in related tables and updates statistics

[russdj]

resolves #391

[csmig]

@russdj . Thanks, looks good. If we end up making more commits to this (see question to Chris below), we might want to make DELETE from stig... the first query. We could therefore remove the explicit DELETE from revision... since the STIG delete will cascade there. But let's see if we want to spend any more time on this PR...

@cd-rite The newman test is very sparse here, just checks for 200. Deleting a STIG initiates cascading deletes that could/should be verified:

• STIG/Asset assignments
• Level 1 access grants

Should we try to get a more robust test into this PR, or delay and make a new issue?

[cd-rite]

I've created a new issue for the tests. #410