Create a list of differences with sbws

to make onbasca behave the same as sbws. After that, we could change some "features".

This issue is to track them.

The differences might affect:

1. External API:

   1.1. Bandwidth File

   • onbasca doesn't under_min_report KeyValue (#45 (closed))
   • onbasca takes the observed bandwidth mean instead of last one (solves sbws#40083 (closed))

   1.2. Logs (from INFO level only?)

   • onbasca rotates by days, no bytes (solves sbws#40066 (closed))

2. Behaviour and/or "Tendence" after some days of an incident (harder to measure)

   2.1. Web Server

   • onbasca doesn't enable/disable the Web Server (#18) (It should use a better algorithm than sbws, cause it was prone to bugs: sbws#40062 (moved), sbws#40067 (closed))

   2.2. Scanner

   • onbasca does not try to download the data 3 times adjusting the size according to the time it took, as sbws does (#2 (closed), #3 (closed)) (It should use a better algorightm than sbws: sbws#30232 (closed), sbws#29291 (moved))
   • onbasca does not measure relays without descriptor (solves sbws#40076 (moved))

   2.3 Generator

   • onbasca scales the bandwidth using last descriptor, not the one available while measureing (solves sbws#40081 (closed))