stake prioritized features for next release

[4meta5]

What does it do?

• use FRAMEv2 macro syntax
• change name to parachain-staking
• remove swap_nominations
• merge nominate_new and join_nominators into single nominate runtime method
• change all instances of Validator to Collator. The Validators storage item should be renamed SelectedCandidates
• add MinCollatorCandidateStk bond, as an associated type for Config trait and only use MinCollatorStk constant when selecting the SelectedCandidates

What important points reviewers should know?

Is there something left for follow-up PRs?

I started to make these changes, but realized they need to be split into follow ups to get proper review.

• make BlocksPerRound a storage item and add runtime method for root to update it (using > instead of % for determining round transitions)
• make collator fee a pallet constant (associated type Get<Perbill> for Config), instead of configurable per collator (as per preferred design constraints)

What alternative implementations were considered?

Are there relevant PRs or issues in other repositories (Substrate, Polkadot, Frontier, Cumulus)?

What value does it bring to the blockchain users?

Checklist

• Does it require a purge of the network?
• You bumped the runtime version if there are breaking changes in the runtime ?
• Does it require changes in documentation/tutorials ?

[4meta5]

If we want, we can set this to be a different value from MinCollatorStk. The new constant is the minimum amount reserved for an account to become a collator candidate. The MinCollatorStk is now the minimum required for a collator candidate to be chosen for the round in SelectedCandidates (which is the set of collators for the round).

Why? If we want to lower the capital requirements for becoming a collator candidate such that accounts can become a candidate with less funds, but they still need to get a certain backing in order to be selected in SelectedCandidates. This was mentioned to me by @artkaseman

[JoshOrndorff]

To me this makes sense in the opposite case of what you described: We want to raise the cost of joining the candidates (because there are too many) but we don't want to boot anyone from the active set (yet).

The case where we lower the cost to become a candidate seems like it might have unintended consequences. Imagine we lower the cost to become a candidate such that it is lower than the total stake to enter the active set. I can image a lot of stakers saying "wth, I made the minimum stake requirement, I got nominators, and I'm in the top N. Why am I not active?"

That being said, if people want it I'm fine with it.

[4meta5]

I can image a lot of stakers saying "wth, I made the minimum stake requirement, I got nominators, and I'm in the top N. Why am I not active?"

If we monitor the chain and make sure that there are at least TotalSelectedCandidates collator candidates with the MinCollatorStk, then this will never happen because anyone in the top N will be put in SelectedCandidates for the round.

[JoshOrndorff]

Looks good. Nice cleanup and update!

I'll mention a few little notes about naming and docs that crossed my mind, but no pressure to make these changes here:

• You've got a notion of SelectedCollators. In the nimbus consensus as well as the research that Gorka has been doing, we've been using the term "Active" where you used Selected. Obviously neither is objectively better.
• One big thing that distinguishes this pallet from Parity's is that this pallet uses direct delegation. Nominators choose exactly who they nominate and with what share of their stake. This is different from Parity's where you approval vote and then run Phragmen. Mightbe worth mentioning in the docs.

[crystalin]

This will require updating the documentation and terms (@albertov19 )