[Bug-Candidate]: Unequal work spread across workers on corpus replay is bottlenecked by worker 0

[0xicingdeath]

Describe the issue:

The logs on startup during corpus replay show that worker 0 is doing the majority of the heavy work (on almost all executions of starting Echidna), with worker 1-15 not doing much, until worker 0 is done loading in its entirety. This seems to be one of the reasons that the corpus replay is so slow.

Code example to reproduce the issue:

https://github.com/curvance/Curvance-CantinaCompetition/tree/CodeFAQAndAdjustments/tests/fuzzing

Version:

Latest Release

Relevant log output:

No response