feat: clarify and stabilize skill memory extraction with controlled chat history usage

[CaralHsi]

Pre-submission checklist | 提交前检查

• I have searched existing issues and this hasn't been mentioned before | 我已搜索现有问题，确认此问题尚未被提及
• I have read the project documentation and confirmed this issue doesn't already exist | 我已阅读项目文档并确认此问题尚未存在
• This issue is specific to MemOS and not a general software issue | 该问题是针对 MemOS 的，而不是一般软件问题

Problem Statement | 问题陈述

当前 Skill Memory 抽取流程在以下方面存在系统性风险：
1. chat_history 使用边界不清晰，可能被误当作主要信息源
2. message index / task chunk 切分在异常输入下不稳定
3. scheduler → reader → processor 的上下文参数传递不完整
4. skill prompt 对“历史上下文”的约束不足，可能导致技能污染
5. 并发与错误处理在高负载下存在潜在问题

期望目标
• 明确 chat_history 的角色，仅作为辅助上下文
• 确保 skill 抽取必须基于当前 messages
• 提升 skill 抽取、切分、写入流程的健壮性
• 为后续 skill 评估、版本化与回溯提供结构保障

The current Skill Memory extraction pipeline has several systemic risks:
1. Unclear boundaries for chat_history usage, allowing it to dominate skill generation
2. Unstable message index and task chunk handling under malformed inputs
3. Incomplete context propagation across scheduler → reader → processor
4. Insufficient prompt constraints on historical context, risking skill pollution
5. Potential concurrency and robustness issues under load

Goals
• Clearly define chat_history as auxiliary context only
• Ensure all extracted skills are grounded in current messages
• Improve robustness of skill extraction, chunking, and persistence
• Lay a solid foundation for future skill evaluation and versioning

Willingness to Implement | 实现意愿

• I'm willing to implement this myself | 我愿意自己解决
• I would like someone else to implement this | 我希望其他人来解决