go: Add consensus transaction prioritization

[abukosek]

This PR adds support for consensus transaction prioritization using the V1 tendermint mempool implementation.

Here are the base priorities for our consensus apps:

consensus/tendermint/apps/beacon/api.go:		AppPriority int64 = 100000
consensus/tendermint/apps/keymanager/api.go:		AppPriority int64 = 50000
consensus/tendermint/apps/registry/api.go:		AppPriority int64 = 50000
consensus/tendermint/apps/governance/api.go:		AppPriority int64 = 25000
consensus/tendermint/apps/roothash/api.go:		AppPriority int64 = 15000
consensus/tendermint/apps/staking/api.go:		AppPriority int64 = 1000
consensus/tendermint/apps/scheduler/api.go:		AppPriority int64 = 0
consensus/tendermint/apps/supplementarysanity/api.go:	AppPriority int64 = 0

The reason why the scheduler and supplementarysanity have a priority of 0 is that they have no transactions.
Node registrations have a slightly higher priority of 60000 instead of the base 50000 used for other registry transactions.

I also experimented a bit with bumping various priorities based on the transaction contents -- e.g. roothash executor commits and proposer timeouts based on runtime stake, but I don't think it's worth the extra complexity.
I suggest we first try the basic version as implemented in this PR and we can always add the extra priority fudging later, once we're sure this works as intended.

In case of errors, the returned priority is 0, which I think makes most sense, so people won't be able to submit faulty transactions to fill up the high-priority part of the queue.

[codecov]

Codecov Report

Merging #4911 (34729a1) into master (06012a8) will decrease coverage by 0.02%.
The diff coverage is 100.00%.

❗ Current head 34729a1 differs from pull request most recent head b844f58. Consider uploading reports for the commit b844f58 to get more accurate results

@@            Coverage Diff             @@
##           master    #4911      +/-   ##
==========================================
- Coverage   66.64%   66.61%   -0.03%     
==========================================
  Files         464      464              
  Lines       51204    51218      +14     
==========================================
- Hits        34127    34121       -6     
- Misses      12891    12904      +13     
- Partials     4186     4193       +7     

Impacted Files	Coverage Δ
go/common/version/version.go	80.26% <ø> (ø)
go/consensus/tendermint/apps/roothash/api.go	100.00% <ø> (ø)
go/consensus/tendermint/abci/mux.go	58.43% <100.00%> (+0.06%)	⬆️
go/consensus/tendermint/api/context.go	88.95% <100.00%> (+0.31%)	⬆️
go/consensus/tendermint/apps/beacon/beacon.go	70.96% <100.00%> (+0.47%)	⬆️
...consensus/tendermint/apps/governance/governance.go	62.50% <100.00%> (-0.54%)	⬇️
...consensus/tendermint/apps/keymanager/keymanager.go	63.24% <100.00%> (+0.19%)	⬆️
go/consensus/tendermint/apps/registry/registry.go	61.94% <100.00%> (+0.57%)	⬆️
go/consensus/tendermint/apps/roothash/roothash.go	73.34% <100.00%> (+0.05%)	⬆️
go/consensus/tendermint/apps/staking/staking.go	57.14% <100.00%> (+0.23%)	⬆️
... and 29 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[kostko]

Can we have some tests that these priorities work correctly?

It would probably be better if these priorities would be emitted similar to how gas stuff is done. E.g. there being some SetPriority in the context that one could query via GetPriority during CheckTx. Then priority could be configured as part of the regular execution flow instead of introducing another method.

[kostko]

Could be moved up into the group where the data, events and gasAccountant are.