R4R Zero Power Block fees to community pool

[rigelrozanski]

Fixes issues seen with seed 99 in jae/simulator_improvements

reference #2906

• Linked to github-issue with discussion and accepted design OR link to spec that describes this work.
• Wrote tests
• Updated relevant documentation (docs/)
• Added entries in PENDING.md with issue #
• rereviewed Files changed in the github PR explorer

---

For Admin Use:

• Added appropriate labels to PR (ex. wip, ready-for-review, docs)
• Reviewers Assigned
• Squashed all commits, uses message "Merge pull request #XYZ: [title]" (coding standards)

[codecov]

Codecov Report

Merging #2905 into jae/simulator_improvements will not change coverage.
The diff coverage is n/a.

@@                     Coverage Diff                     @@
##           jae/simulator_improvements    #2905   +/-   ##
===========================================================
  Coverage                       56.35%   56.35%           
===========================================================
  Files                             120      120           
  Lines                            8416     8416           
===========================================================
  Hits                             4743     4743           
  Misses                           3351     3351           
  Partials                          322      322

[cwgoes]

This does fix this issue, but I don't think this is quite the root of the problem.

I think the problem is that our "LastTotalPower" is not correct because of the one-block delay.

If we unbond all the validators in block n (in the EndBlocker), LastTotalPower will be zero at block n+1 but there will still be validators signing that block since the updates haven't been applied by Tendermint yet.

The simulator's logic is correct in this case, this is what Tendermint would do (although it is very much an edge case we're unlikely to see in real life).

Perhaps the more concerning part is that I think our fee distribution logic really ought to be offset by another block than it actually is at the moment - we're setting accum values based on powers which haven't actually been set by Tendermint yet. Amortized over lots of blocks with few stake distribution changes this seems fine, but there could be edge cases.

[cwgoes]

ACK but see comment, let's think further on this