Wearable Sets are calculated differently in different places (affects Rarity Farming)

[mistaya]

(Full details in my previous reports on discord starting at https://discord.com/channels/732491344970383370/797127275383619604/957269438007218176 and on the forum https://dao.aavegotchi.com/t/wearable-sets-calculation-in-rarity-farming/3878)

When there are several matching wearable sets for a gotchi, the final set chosen is inconsistent in different places.

This affects rarity farming rankings - this is important because the main aavegotchi page in the website can display a different set to the one actually used for the RF leaderboard, misleading owners.

Example: #11267 "569", has "Godlike Stani" set and score 842 on the leaderboard (https://app.aavegotchi.com/leaderboard?page=1), but has the better "Apex Stani" set and score 843 when viewed on its own page (https://app.aavegotchi.com/gotchi/11267).

Final rankings of gotchis can change as a result of the unexpected set, because the chosen set can have a different overall BRS boost, or the set's individual trait boosts can affect the gotchi differently.

I put up my findings here to help explore any issues: https://www.aadventure.io/rf/3/2 (enable the debugging option in the "Advanced table config" to show the extra columns).

Suggested fix:

• Decide upon an algorithm for picking the best wearable set. This needs to have multiple levels of tiebreaker, with a final fallback that is guaranteed to produce a consistent order.
• Publish the algorithm, so that community devs can also implement it consistently in their tools.
• Implement the algorithm in the main website and leaderboard calculations (maybe also fix the core-matic subgraph, or just remove the inaccurate equippedSetID, equippedSetName, withSetsRarityScore properties to avoid misleading community devs).

Example algorithm:

1. highest total set bonus (BRS plus magnitude of NRG, AGG, SPK, BRN bonuses)
2. highest number of wearables in the set
3. index of when the set was added onchain (earliest wins)