Fix stuck post init on no VRF nonce

[fasmat]

Closes #316

The problem is that the default ComputeBatchSize is just too large for small inits on testnet or fasnat presets. To avoid an initialization that didn't find a VRF nonce to run much longer than it needs to when searching for a nonce outside of the initialization range the actual used ComputeBatchSize is limited to the total size of the initialization. Larger values don't make sense, since a nonce should statistically be found with ~ 99.97% probability during initialization every subsequent batch of the same size as the PoST should have another 99.97% of containing a valid nonce. Failing to find a nonce more than 2x outside of the init range is statistically impossible, so there is no need to have a batch size bigger than the total init size 🙂

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 59.9%. Comparing base (3cf63ad) to head (582280b).
Report is 2 commits behind head on develop.

Additional details and impacted files

@@           Coverage Diff           @@
##           develop    #317   +/-   ##
=======================================
  Coverage     59.8%   59.9%           
=======================================
  Files           30      30           
  Lines         2291    2294    +3     
=======================================
+ Hits          1372    1375    +3     
  Misses         781     781           
  Partials       138     138           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.