feat: Improve MultiModalStructMemReader: Better Language Detection, Prompt Selection, and Source-Aware Processing

[CaralHsi]

Description

Summary: This PR enhances the MultiModalStructMemReader to more accurately detect languages, choose the appropriate prompt type, and leverage source content when constructing LLM requests. It also improves robustness and keeps backward compatibility with existing pipeline behavior.

1. Source-aware language detection
2. Automatic prompt type selection (chat vs doc)
3. Enhanced _get_llm_response()

Fix: #(issue)

Docs Issue/PR: (docs-issue-or-pr-link)

Reviewer: @(reviewer)

Checklist:

• I have performed a self-review of my own code | 我已自行检查了自己的代码
• I have commented my code in hard-to-understand areas | 我已在难以理解的地方对代码进行了注释
• I have added tests that prove my fix is effective or that my feature works | 我已添加测试以证明我的修复有效或功能正常
• I have created related documentation issue/PR in MemOS-Docs (if applicable) | 我已在 MemOS-Docs 中创建了相关的文档 issue/PR（如果适用）
• I have linked the issue to this PR (if applicable) | 我已将 issue 链接到此 PR（如果适用）
• I have mentioned the person who will review this PR | 我已提及将审查此 PR 的人