Calculate weighted severity, exploitability and risk score for advisories and packages at the time of import as well

Right now we only have an improver to do it for both advisories and packages but we should calculate risk_score, weighted_severity and risk score for advisories at time of import with avialbale info at that time and keep improving the data if we get new info on it.

For packages as we should get risk score at import time and keep improving the data with new improvements