Cooldown changes

[td202]

Purpose

Closes #1142 #1145

Implement cooldown changes in the node.

Changes

• Split bsoConfigureBaker into bsoAddValidator and bsoUpdateValidator. Rewrite to factor out the checks and to make the flow of account updates more apparent.
  • The behaviour of bsoUpdateValidator is changed in the following case: when the validator's keys are changed, and the new aggregation key duplicates the aggregation key of some other validator, the key that is actually duplicated is now returned (whereas previously the old key on the account was returned) in the error. This causes a difference in the RejectReason which is reported for such a transaction, and could cause incompatibility if such a difference was observed in P1-P4 (since P5, reject reasons are not included in the block hashing scheme). (This does not seem to have occurred at least on mainnet.)
• Split bsoConfigureDelegation into bsoAddDelegator and bsoUpdateDelegator. Rewrite to factor out the checks and to make the flow of account updates more apparent.
• Extensive tests. To assure backwards compatibility (except in the above noted difference), the tests were also run against the old bsoConfigureBaker and bsoConfigureDelegation (see the branch: https://github.com/Concordium/concordium-node/tree/backport-tests).

Checklist

• My code follows the style of this project.
• The code compiles without warnings.
• I have performed a self-review of the changes.
• I have documented my code, in particular the intent of the
  hard-to-understand areas.
• (If necessary) I have updated the CHANGELOG.

[Victor-N-Suadicani]

I pushed a fix to the consensus, because it didn't actually recognize the P7 protocol update. I'm not really sure why I didn't see this when I was testing the P6 to P7 update, where it seemed to work fine, but for P7 to P7 it did not work. In any case, it should be fixed now. I also changed it from using a wildcard pattern match to avoid future problems like this.

I tested the update with this fix and tried the (newly updated) state compare tool as well on the protocol update and it seemed to work perfectly fine.