Extend desirability function to consider oversaturated pools

[ilap]

Desirability function is based on the assumption of the Nash equilibrium, and therefore does not consider the possibility of the over-saturation of the pools.
As a consequence, an over-saturated pool will have the same desirability, and therefore ranking, than any non- or fully saturated pool.

Though, I understand the logic of the decision, but I think the desirability and therefore the ranking should consider the over-saturated pools to indirectly incentivase the users to delegate to the non-saturated pools.

This new addition could be considered as some max threshold i.e. max 100% of the desirability, which depends only on the level of the oversaturation of the relevant pool.

This can be very easily implemented in the haskell code by slightly modifying these three function belows:

• https://github.com/input-output-hk/cardano-ledger-specs/blob/9dd938b2fa66c0712293aa7f273ca9450bdd05a6/shelley/chain-and-ledger/executable-spec/src/Shelley/Spec/Ledger/Rewards.hs#L251
• https://github.com/input-output-hk/cardano-ledger-specs/blob/9dd938b2fa66c0712293aa7f273ca9450bdd05a6/shelley/chain-and-ledger/executable-spec/src/Shelley/Spec/Ledger/Rewards.hs#L277
• https://github.com/input-output-hk/cardano-ledger-specs/blob/9dd938b2fa66c0712293aa7f273ca9450bdd05a6/shelley/chain-and-ledger/executable-spec/src/Shelley/Spec/Ledger/API/Wallet.hs#L60

[ilap]

I do not think it is necessary as the non-myopic reward would consider the oversaturated stake later anyway.
But, it would be a nice to have feature for checking the oversaturation at the ranking/desirability part too and not just in the non-myopic reward, to allow some others to be fit in the k number of pools instead of an over-saturated one.

[nc6]

I think the sensible thing here is to follow the CIP: https://github.com/cardano-foundation/CIPs/blob/master/CIP1/CIP1.md

The current ranking scheme is operating as designed; a change to the design needs to be proposed and considered more widely. Issues here are more for things we know we want to change!