v0.81.1

Changed

• draugr feeds update keeps a cached feed when the fetch fails, instead of failing the run. A
  403 from CISA blocked a pipeline that already had the catalogue on disk. The step exists so a
  scan cannot rank everything as though nothing were exploited — and a cached catalogue does not do
  that: it ranks on data of a known age, and the report says how old. The failed fetch is reported
  with the age of the copy it kept, and warned about, so a run whose ranking is older than you
  think says so. With nothing cached there is no answer to keep, and it still fails.