log: retract v7-v12 KL conclusion; metric artifact + logp_floor fix

log: repetition fix necessary but not sufficient

log: v7-v12 stability campaign, 3 bugs fixed, no strength gain

log: repetition-feature bug, greedy baseline, learn bench

fix eval-on-resume (silently skipped); add learn bench; v4 analysis

add learn-phase thread benchmark

revert logp_old change (ablation-falsified); keep only_player + kl estimator fixes

draw-penalty: remove guaranteed-0 repetition optimum

parallel self-play (--workers), conversion experiments, whitelist

whitelist analysis scripts; fix pid-recycling in launch/job_table

research log: env correction + thread bench

tinychess: full research artifact - self-play RL, ablations, compute scaling, faithfulness, precision, plasticity