feat: add reranker

[CaralHsi]

Description

Summary:

This PR introduces a pluggable Reranker layer to MemOS, wires it into Tree Textual Memory and MOS product configs, and tweaks memory filtering thresholds. It adds two backends out-of-the-box—cosine_local (uses embeddings) and http_bge (OpenAI-style rerank API)—plus an example script and configuration factories.
GitHub

• New reranker module & factory:
  Adds memos/reranker/ with base.py, cosine_local.py, http_bge.py, noop.py, and factory.py to construct rerankers via config.

• Config plumbing
  New RerankerConfigFactory (src/memos/configs/reranker.py) to describe backend & options.

• TreeTextMemoryConfig gains optional reranker and memory_size fields so tree memories can specify reranker and bucket caps.

• APIConfig: adds get_reranker_config() and injects it into both create_user_config() and default cube config; supports env-driven selection (MOS_RERANKER_BACKEND, MOS_RERANKER_URL). Defaults to http_bge else cosine_local.

• Product behavior
  Slightly raises retrieval acceptance threshold and removes minimum count: threshold 0.50 → 0.52, min_num 3 → 0.

• Examples & tests
  Adds examples/basic_modules/reranker.py demonstrating both backends; deletes older tree_textual_memory_reranker.py.
  Updates tree searcher test (tests/memories/textual/test_tree_searcher.py) accordingly.

Fix: #(issue)

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

Reviewer: @fridayL

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