revert(levels): ship round 2's set — contrast selection measurably made it easier

Round 3 tested one hypothesis from round 2's data: levels the agent rated 3
carried 59% more 反差 than the rest, so select for it. With both rounds now at
~100 runs, the answer is in and it is no. The back band went from 2.39 ± 0.10 to
2.19 ± 0.08, roughly two standard errors apart, and the levels averaging 2.5 or
better fell from 11 to 6. Contrast is real and it does correlate with being
rated harder; selecting on it does not produce harder levels.

The low retest reliability does not rescue the change either — r = 0.55 limits
ranking one level against another, not the mean of eighteen levels measured a
hundred times.

So round 2's levels are the ones handed over, rebuilt from its runs via
round-set --emitCampaign. select-50 cannot reproduce them: the pool it drew
from has been rebuilt since, so the same flags now return different levels. The
runs are the record.

Rule bot on the shipped set: 50/50 cleared, dead taps 0.00, blind rotations
0.12, worst blind streak 1. 198 tests pass.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

docs/level-board.html

@@ -110,7 +110,7 @@ footer{margin-top:56px;padding-top:20px;border-top:1px solid var(--rule);color:v
   <div class="tile ok"><div class="k">候选关卡</div><div class="v">50 <span style="font-size:14px;color:var(--ink3)">/ 50</span></div><div class="n">当前关卡集</div></div>
   <div class="tile ok"><div class="k">盲转 / 关</div><div class="v">0.00</div><div class="n">目标 0</div></div>
   <div class="tile ok"><div class="k">转错方向 / 关</div><div class="v">0.00</div><div class="n">目标 0</div></div>
-  <div class="tile ok"><div class="k">猴子 动作÷棋子</div><div class="v">6.11</div><div class="n">目标 ≥3.0（不无脑）</div></div>
+  <div class="tile ok"><div class="k">猴子 动作÷棋子</div><div class="v">6.43</div><div class="n">目标 ≥3.0（不无脑）</div></div>
 </div>
 
 <div class="scroll"><table>
@@ -118,12 +118,12 @@ footer{margin-top:56px;padding-top:20px;border-top:1px solid var(--rule);color:v
 <tbody>
 <tr><td>1 转</td><td class="n">5</td><td class="sub">浅</td></tr>
 <tr><td>2 转</td><td class="n">5</td><td class="sub">浅</td></tr>
-<tr><td>3 转</td><td class="n">26</td><td class="sub">浅</td></tr>
-<tr><td>4 转</td><td class="n">10</td><td class="sub">中</td></tr>
+<tr><td>3 转</td><td class="n">24</td><td class="sub">浅</td></tr>
+<tr><td>4 转</td><td class="n">12</td><td class="sub">中</td></tr>
 <tr><td>5 转</td><td class="n">4</td><td class="sub">中</td></tr>
 
 </tbody></table></div>
-<p class="sub">路线家族分布：v05 ×10 · spiral ×5 · switchback ×13 · orbits ×9 · weave ×13　终局可走朝向分布：{"3":44,"4":6}</p>
+<p class="sub">路线家族分布：v05 ×10 · spiral ×2 · switchback ×12 · weave ×18 · orbits ×8　终局可走朝向分布：{"3":44,"4":5,"5":1}</p>
 
 <h2><span class="num">B</span>什么叫「难度合适且好玩」</h2>
 <div class="callout"><p class="sub">旋转是免费的、是这个游戏的特色，玩家应该频繁做——所以不该限制旋转次数。该限制的是每一次旋转值不值：转了有没有棋子出来、能不能看出来该转哪、以及转完之后有没有「原来这么简单」的反差。评分里只有 difficulty（复测 r=0.93）和 stuck（r=0.85）可信，rotation / clarity / fun 是噪声（0.32 / 0.19 / 0.56），所以「好玩」全部换成机械代理。 另外，演出本身就是奖励——棋子沿弯路走出去的那几秒是减压点，所以开局必须有得看，而不是只有两枚贴边棋子和 0.3 秒。</p></div>
@@ -184,14 +184,14 @@ footer{margin-top:56px;padding-top:20px;border-top:1px solid var(--rule);color:v
   <li class="done"><div class="t"><span class="tid">T3.1</span><span class="tt">第 1 轮：grok 全量跑 50 关</span><span class="tag good">完成</span></div><div class="note">50 关全部跑完；机械面干净：空点 0、盲转 ~0.3/关、全部通关。总体读数：整批偏易一档（2 分 vs 刚好的 3 分）</div></li>
 <li class="done"><div class="t"><span class="tid">T3.2</span><span class="tt">按实测曲线调整选关，重选</span><span class="tag good">完成</span></div><div class="note">连续空转设闸门（≤1）+ 难度曲线改为按旋转深度配额。同一规则机器人对照：盲转 0.24→0.12，最长连续盲转 4→1，决策数不变</div></li>
 <li class="done"><div class="t"><span class="tid">T3.3</span><span class="tt">第 2 轮验证</span><span class="tag good">完成</span></div><div class="note">R2：99 局覆盖 50 关，全部通关，空点 0。难度按档 2.00 / 2.05 / 2.40，无解感 1.21 / 1.15 / 1.74 —— 评分第一次出现梯度</div></li>
-<li class="done"><div class="t"><span class="tid">T3.4</span><span class="tt">收敛判定（难度曲线单调、stuck 达标）</span><span class="tag good">完成</span></div><div class="note">收敛判定完成。机械面：空点 0、不可解态 0、连续空转≤1、终局朝向≥3、三轮全部通关。难度面：三档 2.00 / 2.05 / 2.17，单调但后档低于「刚刚好」的 3 分——上限由深关卡供给决定（见 T1.6），且这项评分复测 r=0.55，本身不足以再往下调。剩下的交人工审核</div></li>
+<li class="done"><div class="t"><span class="tid">T3.4</span><span class="tt">收敛判定（难度曲线单调、stuck 达标）</span><span class="tag good">完成</span></div><div class="note">收敛判定完成。机械面全部达标：空点 0、不可解态 0、连续空转≤1、终局朝向≥3、三轮全部通关。难度面 2.00/2.05/2.39 单调，后档仍低于「刚刚好」的 3 分，上限由深关卡供给决定（T1.6）。第三轮试图用反差再推一把，被数据否掉并回退</div></li>
   </ul>
 </div>
 
 <div class="phase">
   <h3><span class="pid">P4</span>交付</h3>
   <ul class="tasks">
-  <li class="done"><div class="t"><span class="tid">T4.1</span><span class="tt">50 关关卡集</span><span class="tag good">完成</span></div><div class="note">levels/campaign-50.json + src/levels-c50.js，已注册为可玩版本 c50，198 项测试全绿</div></li>
+  <li class="done"><div class="t"><span class="tid">T4.1</span><span class="tt">50 关关卡集</span><span class="tag good">完成</span></div><div class="note">交付的是第二版（R2 跑的那一版），从其 run 反建，50 关旋转深度 {1:5,2:5,3:24,4:12,5:4}，家族 weave18/switchback12/v05 10/orbits8/spiral2。198 项测试全绿，规则机器人复核 50/50 通关、空点 0、最长连续盲转 1</div></li>
 <li class="done"><div class="t"><span class="tid">T4.2</span><span class="tt">人工审核页（可直接试玩每一关）</span><span class="tag good">完成</span></div><div class="note">docs/review-50.html：50 张开局盘面缩略图 + 机械指标 + agent 实测 + 试玩直达；并标注「哪些数字能信」，把感知类问题（路线长度、紧贴）显式交给人眼</div></li>
 <li class="done"><div class="t"><span class="tid">T4.3</span><span class="tt">后续关卡制作流程的优化总结</span><span class="tag good">完成</span></div><div class="note">docs/level-process.html：两轮对照、复测可信度表（本批现算）、五条规则各带证据、被否掉的做法清单</div></li>
   </ul>
@@ -204,8 +204,8 @@ footer{margin-top:56px;padding-top:20px;border-top:1px solid var(--rule);color:v
 <div class="scroll"><table>
 <thead><tr><th>批次</th><th class="n">关数</th><th class="n">difficulty</th><th class="n">stuck</th><th class="n">盲转/关</th><th class="n">多转/关</th></tr></thead>
 <tbody>
+<tr><td>R3 · grok xhigh</td><td class="n">100</td><td class="n">2.10</td><td class="n">1.35</td><td class="n">0.14</td><td class="n">0.26</td></tr>
 <tr><td>R2 · grok xhigh</td><td class="n">100</td><td class="n">2.16</td><td class="n">1.37</td><td class="n">0.12</td><td class="n">0.24</td></tr>
-<tr><td>R3 · grok xhigh</td><td class="n">53</td><td class="n">2.08</td><td class="n">1.32</td><td class="n">0.13</td><td class="n">0.25</td></tr>
 <tr><td>R1 · grok xhigh</td><td class="n">50</td><td class="n">2.10</td><td class="n">1.44</td><td class="n">0.24</td><td class="n">0.38</td></tr>
 <tr><td>Codex CLI · gpt-6-astra low</td><td class="n">18</td><td class="n">2.44</td><td class="n">1.44</td><td class="n">1.17</td><td class="n">1.22</td></tr>
 <tr><td>Codex CLI · spark xhigh</td><td class="n">18</td><td class="n">3.11</td><td class="n">2.28</td><td class="n">1.17</td><td class="n">1.17</td></tr>
@@ -221,6 +221,7 @@ footer{margin-top:56px;padding-top:20px;border-top:1px solid var(--rule);color:v
 
 <h2><span class="num">F</span>日志</h2>
 <ul class="log">
+<li><b>第三轮回退</b>按反差选关后，后档评分 2.39±0.10 → 2.19±0.08（约两个标准误），平均分≥2.5 的关从 11 掉到 6。反差确实和「被评得更难」相关，但按它选关造不出更难的关——相关不等于可操纵。已回退到第二版</li>
 <li><b>一个差点报错的对照</b>levels/campaign-50-r2.json 本应是 R2 玩的那一版，实际只含 28 个 R2 盘面、46 个 R3 盘面——select-50 原地覆盖同一路径，中间跑过的选关实验把它换掉了。改为从 run 反建（run 自带完整盘面与序号），并删掉该快照</li>
 <li><b>R3 结果</b>第三轮只改一个变量：桶内按反差取关。后档反差 27.9→28.5、旋转深度不变，后档评分 2.39→2.17。以 r=0.55 的复测可信度，十几关上的这点差距分不出胜负——记为「未证实」，第三版保留的理由是设计面的（反差是策划点名的好玩机制）与家族均衡（两家族 30/50 → 26/50）</li>
 <li><b>供给瓶颈</b>扩池两轮共 720 种子、1113 候选，深度≥4 的过闸门后只有 23 个（2%）。种子翻倍只让深关卡从 16 涨到 23，说明 growDeepLevel 难以造出需要四转以上的关卡——这是后档难度的硬上限</li>