Renumber the stages: ground truth is V3.1, released as v3.1.0

The owner decided that the combat ground-truth stage, first planned as
V4, mainly adds a research foundation and a Genichiro content update to
the V3 product, so it becomes V3.1 and ships as v3.1.0. The later
stages move down by one: automated gameplay analysis is now V4 and the
multi-game platform V5, so stage names and release numbers line up.

CLAUDE.md, both roadmaps, both READMEs and the research docs use the
new names; CLAUDE.md and the roadmaps explain the change for readers of
older commits and PR #23. The status sections now say V3.1 Milestones
1-5 are done and product integration is current, and the Genichiro
plan's progress and todo list are brought up to date.

Code identifiers and the wiki mapping's v4_status field keep their
names.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Author: CanYouTeachMeHowToCode

CLAUDE.md

@@ -181,7 +181,7 @@ V2 runs locally with Docker Compose.
 
 #### Deferred or Skipped
 
-* **Public deployment** was deferred: without accounts, a public instance would let anyone record attempts. V3 adds accounts. Public hosting is now planned for V5, once a player can upload a recording and see which moves caused their failures.
+* **Public deployment** was deferred: without accounts, a public instance would let anyone record attempts. V3 adds accounts. Public hosting is now planned for V4, once a player can upload a recording and see which moves caused their failures.
 * **Boss search and filtering** was skipped, because 8 bosses fit comfortably on one page.
 
 ---
@@ -196,10 +196,10 @@ V3 turned the single-user local app into a foundation that several players can u
 * ✅ **User-owned attempts:** every attempt belongs to the player who recorded it. Attempts recorded before accounts existed are kept and can be claimed with `scripts.claim_attempts`.
 * ✅ **Per-user analytics:** attempt history, boss analytics, progression, and the Sekiro dashboard only cover your own attempts.
 * ✅ **English / 中文 interface:** a language switch, with each language showing only its own text. The choice is saved on the account, or in the browser for visitors. Most Chinese move names and descriptions are translations of the English source material.
-* ✅ **Existing content cleanup:** the current boss data was checked against two wikis (Fextralife and Fandom); errors and misleading text were fixed and phase structures corrected, rather than adding more bosses from wikis. The comparison for every boss is kept in [`research/ground_truth/semantic/wiki_mapping/`](research/ground_truth/semantic/wiki_mapping/README.md) as the starting point for V4.
+* ✅ **Existing content cleanup:** the current boss data was checked against two wikis (Fextralife and Fandom); errors and misleading text were fixed and phase structures corrected, rather than adding more bosses from wikis. The comparison for every boss is kept in [`research/ground_truth/semantic/wiki_mapping/`](research/ground_truth/semantic/wiki_mapping/README.md) as the starting point for V3.1.
 * ✅ **Hardening:** the four required CI jobs also check that accounts and attempts survive a Docker restart.
 
-While expanding boss coverage, community wiki movesets turned out to be incomplete and sometimes inconsistent with each other. V3 therefore treats boss data as human-readable reference content, not as a complete record of every attack in the game. Verifying movesets against the game itself is the job of V4.
+While expanding boss coverage, community wiki movesets turned out to be incomplete and sometimes inconsistent with each other. V3 therefore treats boss data as human-readable reference content, not as a complete record of every attack in the game. Verifying movesets against the game itself is the job of V3.1.
 
 ---
 
@@ -208,17 +208,17 @@ While expanding boss coverage, community wiki movesets turned out to be incomple
 The full plan is in [ROADMAP.md](ROADMAP.md). In short:
 
 ```text
-V1  What killed me?                      → manual attempt analytics
-V2  Where am I improving or struggling?  → structured relational analytics
-V3  Can different players use it?        → multi-user product foundation
-V4  What moves actually exist?           → combat ground truth
-V5  What happened in the fight?          → gameplay video analysis + public deployment
-V6  Can this work across games?          → multi-game platform
+V1    What killed me?                      → manual attempt analytics
+V2    Where am I improving or struggling?  → structured relational analytics
+V3    Can different players use it?        → multi-user product foundation
+V3.1  What moves actually exist?           → combat ground truth
+V4    What happened in the fight?          → gameplay video analysis + public deployment
+V5    Can this work across games?          → multi-game platform
 ```
 
-* **V4** builds verified boss data from the game itself, through runtime instrumentation and targeted reverse engineering, starting with Genichiro.
-* **V5** analyzes uploaded gameplay, starting with the failures in a recording and the moves that caused them; the app is deployed publicly once that works. It then detects every boss move and the player's responses. Knowing how often each move actually occurred makes true per-move success rates and practice recommendations possible.
-* **V6** extends the platform beyond Sekiro, starting with Black Myth: Wukong.
+* **V3.1** builds verified boss data from the game itself, through runtime instrumentation and targeted reverse engineering, starting with Genichiro.
+* **V4** analyzes uploaded gameplay, starting with the failures in a recording and the moves that caused them; the app is deployed publicly once that works. It then detects every boss move and the player's responses. Knowing how often each move actually occurred makes true per-move success rates and practice recommendations possible.
+* **V5** extends the platform beyond Sekiro, starting with Black Myth: Wukong.
 
 The long-term goal is to turn gameplay footage into trustworthy, explainable boss-practice analytics with as little manual input as possible.
 

README.md

@@ -181,7 +181,7 @@ V2 使用 Docker Compose 在本地运行。
 
 #### 推迟或跳过的内容
 
-* **公网部署**被推迟了：在没有账号系统的情况下，公开的实例会让任何人都能新增攻略记录。V3 加入了账号系统；公网部署现在计划放在 V5，等玩家上传录像后能看到是哪些招式导致了失误，再进行。
+* **公网部署**被推迟了：在没有账号系统的情况下，公开的实例会让任何人都能新增攻略记录。V3 加入了账号系统；公网部署现在计划放在 V4，等玩家上传录像后能看到是哪些招式导致了失误，再进行。
 * **Boss 搜索和筛选**被跳过了，因为 8 个 Boss 在一个页面上就能放得下。
 
 ---
@@ -196,10 +196,10 @@ V3 把单用户的本地应用，变成可以供多名玩家使用、每人数
 * ✅ **攻略记录归属用户：** 每条记录都属于记录它的玩家。有账号系统之前的旧记录都保留着，可以用 `scripts.claim_attempts` 认领。
 * ✅ **按用户的分析：** 攻略历史、Boss 分析、进度图和只狼总览都只统计你自己的记录。
 * ✅ **中英文界面：** 可以切换语言，每种语言只显示该语言的文字。登录用户的选择保存在账号里，访客的选择保存在浏览器里。大部分中文招式名和说明译自英文资料。
-* ✅ **清理现有内容：** 对照两个 Wiki（Fextralife 和 Fandom）检查了现有的 Boss 数据，修正了错误和有误导性的描述，也修正了阶段结构，没有继续从 Wiki 补充更多 Boss。每个 Boss 的对照结果保存在 [`research/ground_truth/semantic/wiki_mapping/`](research/ground_truth/semantic/wiki_mapping/README.md)，作为 V4 的起点。
+* ✅ **清理现有内容：** 对照两个 Wiki（Fextralife 和 Fandom）检查了现有的 Boss 数据，修正了错误和有误导性的描述，也修正了阶段结构，没有继续从 Wiki 补充更多 Boss。每个 Boss 的对照结果保存在 [`research/ground_truth/semantic/wiki_mapping/`](research/ground_truth/semantic/wiki_mapping/README.md)，作为 V3.1 的起点。
 * ✅ **加固：** 4 个必须通过的 CI 任务还会检查重启 Docker 后账号和攻略记录是否仍然保留。
 
-在扩充 Boss 数据的过程中，我们发现社区 Wiki 的招式资料并不完整，不同 Wiki 之间有时还互相矛盾。所以 V3 把 Boss 数据定位为方便阅读的参考资料，而不是游戏里全部攻击的完整记录。用游戏本身验证招式数据，是 V4 的工作。
+在扩充 Boss 数据的过程中，我们发现社区 Wiki 的招式资料并不完整，不同 Wiki 之间有时还互相矛盾。所以 V3 把 Boss 数据定位为方便阅读的参考资料，而不是游戏里全部攻击的完整记录。用游戏本身验证招式数据，是 V3.1 的工作。
 
 ---
 
@@ -208,17 +208,17 @@ V3 把单用户的本地应用，变成可以供多名玩家使用、每人数
 完整计划见 [ROADMAP.zh-CN.md](ROADMAP.zh-CN.md)（[English](ROADMAP.md)）。简要来说：
 
 ```text
-V1  是什么打败了我？         → 手动记录的尝试分析
-V2  我在哪里进步、在哪里卡住？ → 结构化的关系型分析
-V3  不同玩家都能用吗？       → 多用户产品基础
-V4  到底有哪些招式？         → 战斗真实数据
-V5  这场战斗里发生了什么？   → 实战录像分析 + 公网部署
-V6  能用于其他游戏吗？       → 多游戏平台
+V1    是什么打败了我？         → 手动记录的尝试分析
+V2    我在哪里进步、在哪里卡住？ → 结构化的关系型分析
+V3    不同玩家都能用吗？       → 多用户产品基础
+V3.1  到底有哪些招式？         → 战斗真实数据
+V4    这场战斗里发生了什么？   → 实战录像分析 + 公网部署
+V5    能用于其他游戏吗？       → 多游戏平台
 ```
 
-* **V4** 通过运行时监测和有针对性的逆向工程，从游戏本身建立经过验证的 Boss 数据，先从苇名弦一郎开始。
-* **V5** 分析上传的实战录像，先找出录像里的失误以及导致失误的招式；这一步跑通后进行公网部署。之后再识别全部 Boss 招式和玩家的应对。知道每个招式实际出现了多少次，才能算出真正的招式成功率，并给出练习建议。
-* **V6** 把平台扩展到只狼以外的游戏，第一个是黑神话：悟空。
+* **V3.1** 通过运行时监测和有针对性的逆向工程，从游戏本身建立经过验证的 Boss 数据，先从苇名弦一郎开始。
+* **V4** 分析上传的实战录像，先找出录像里的失误以及导致失误的招式；这一步跑通后进行公网部署。之后再识别全部 Boss 招式和玩家的应对。知道每个招式实际出现了多少次，才能算出真正的招式成功率，并给出练习建议。
+* **V5** 把平台扩展到只狼以外的游戏，第一个是黑神话：悟空。
 
 长期目标是：把实战录像变成可信、可解释的 Boss 练习分析，并尽量减少玩家的手动输入。