Remove MinDelegatorStk in favor of MinDelegation pallet-staking const

[nbaztec]

What does it do?

removes MinDelegatorStk in favor of MinDelegation pallet-staking const.

On almost all our networks the values were same and therefore redundant. Additionally this caused additional complexity/odd behavior during revoke, where for delegations D1 with a MinDelegation (= 3) bond and D2 with a higher MinDelegatorStk bond (= 5), D2 would not be able to schedule a revoke since the delegators total bond would be below the MinDelegatorStk (5)

This was previously circumvented via the schedule_leave_delegators extrinsic, that was deprecated and is set to be removed.

This PR simplifies the process by removing the MinDelegatorStk and simply using the MinDelegation as the sole condition for a delegation to be instantiated, which by definition makes the user a delegator.

tl;dr A user is a delegator if it has atleast 1 delegation with a bond value of MinDelegation or more.

⚠️ Breaking Changes ⚠️

❗ Removes MinDelegatorStk from pallet-parachain-staking config, in favor of MinDelegation

❗ The extrinsics delegate and delegateWithAutoCompound now return DelegationBelowMin error instead of the former DelegatorBondBelowMin when being called with a value less than MinDelegation

[notlesh]

LGTM.

I know that we had this deployed in most cases as MinDelegatorStk == MinDelegation, but are there any concerns for runtimes where this isn't the case?

I tried to do some basic analysis of this, but it gets complicated because there are potentially checks across runtimes (e.g. you could schedule a request in runtime X and execute it in runtime X+1). It looks to me like such cases would leave an unexecutable request in storage because it will always error when executed. That said, this is probably no different than adjusting the constant between runtimes...

[github-actions]

Coverage generated "Fri Jul 7 13:00:25 UTC 2023":
https://s3.amazonaws.com/moonbeam-coverage/pulls/2381/html/index.html

Master coverage: 70.94%
Pull coverage: 70.92%