[RL] [KVCache] let cache transfer managers update key prefix after weight update and add unit tests

[liyonghua0910]

Motivation

This PR updates the mooncake key_prefix during update_weights.

Previously, key_prefix was only refreshed in resume. To make cache metadata align with the new model version earlier, this PR adds one more refresh in update_weights, while keeping the existing refresh in resume.

Modifications

• Add cache-transfer control handling in EngineService._control_update_weights()
• Add update_weights handler in CacheTransferManager
• Keep resume refreshing key_prefix
• Add unit tests for both update_weights and resume paths

Usage or Command

No new API.

Behavior after this PR:

• update_weights refreshes key_prefix
• resume also refreshes key_prefix

pytest -q tests/engine/test_common_engine.py -k "control_update_weights"
pytest -q tests/cache_manager/test_cache_transfer_manager.py -k "handle_resume or handle_update_weights"

Accuracy Tests

N/A

Checklist

• Add at least a tag in the PR title.
• Format your code, run pre-commit before commit.
• Add unit tests.
• Provide accuracy results.
• If the current PR is submitting to the release branch, make sure the PR has been submitted to the develop branch, then cherry-pick it to the release branch with the [Cherry-Pick] PR tag.

[paddle-bot]

Thanks for your contribution!

[codecov-commenter]

Codecov Report

❌ Patch coverage is 52.94118% with 8 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@3651113). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
fastdeploy/engine/common_engine.py	27.27%	7 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #7083   +/-   ##
==========================================
  Coverage           ?   74.15%           
==========================================
  Files              ?      376           
  Lines              ?    52873           
  Branches           ?     8251           
==========================================
  Hits               ?    39207           
  Misses             ?    10908           
  Partials           ?     2758           

Flag	Coverage Δ
GPU	74.15% <52.94%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[fastdeploy-bot]

🤖 AI Code Review | {{TIMESTAMP}}

📋 Review 摘要

PR 概述：在 update_weights 操作中添加缓存 key_prefix 的更新，使缓存元数据与新模型版本更早保持一致
变更范围：cache_manager/、engine/、测试文件
影响面 Tag：KVCache Engine

问题

未发现阻塞性问题。

代码亮点

1. request_id 改进：将硬编码的 pause_transfer/resume_transfer 改为基于父请求 ID 的命名方式（如 {control_request.request_id}_pause_transfer），提升了请求追踪的可调试性。

2. 一致的条件检查：_handle_update_weights 和 _handle_resume 都正确检查了 storage_backend_type is not None 才调用 _update_key_prefix()。

3. 测试覆盖充分：新增测试覆盖了正常路径（有 storage backend）和跳过路径（无 storage backend），以及相关的控制方法行为。

总体评价

代码实现清晰，逻辑正确，测试覆盖充分。PR 标题和描述均符合规范，可以合入。