LIP-36: update spec

[kyriediculous]

• add LIP_UPGRADE_ROUNDS map and setLIPUpgradeRound(_LIP, _round) details

• update rewards algorithm

• update fees algorithm

• update claimEarnings algorithm

[kyriediculous]

8d1ccf8 fixes some typo's and spacing.

[kyriediculous]

One thing to note is that using an LIP_36_UPGRADE_ROUND as we do now means that users who call reward() in that round prior to the upgrade will lose out on their rewards for that round (as LIP-36 is currently implemented) , I feel like this is a reasonable tradeoff to make vs having to set a state variable on the earningsPool for each round. We can communicate this to alleviate the concerns. This small discrepancy would exist in the snapshot approach as well regardless so I feel like it's okay if it's unilaterally across both "transitional options" (albeit that the snapshot is still very much just in ideation).