[Cherry-Pick][RL] Reuse GDR checkpoint transfer handle(#8078)

[jackyYang6]

Motivation

Avoid repeated CheckpointTransfer initialization during GDR dynamic weight updates. Reusing the initialized handle reduces repeated setup overhead across multiple update steps.

Modifications

• Cache the GDR CheckpointTransfer handle in DynamicWeightManager.
• Lazily initialize the handle on the first GDR weight update.
• Reuse the cached handle for later update_weights_by_gdr calls.
• Destroy and reset the cached handle when an update fails.

Usage or Command

No new user-facing command. Existing GDR weight update flow is unchanged.

Accuracy Tests

Not applicable. This PR only changes checkpoint-transfer handle initialization behavior and does not affect model outputs.

Checklist

• Add at least a tag in the PR title.
• Format your code, run pre-commit before commit.
• Add unit tests. No unit tests added because this is a handle lifecycle optimization for GDR runtime behavior.
• Provide accuracy results. Not applicable; no model output changes.
• 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.

[PaddlePaddle-bot]

🤖 Paddle-CI-Agent | pr_review | 2026-06-25 20:03:04

📋 Review 摘要

PR 概述：为 RL 动态权重更新的 GDR 路径缓存并复用 CheckpointTransfer handle。
变更范围：fastdeploy/rl/dynamic_weight_manager.py、tests/rl/test_dynamic_weight_gdr.py
影响面 Tag：[RL]

问题

未发现阻塞性问题。PR 规范问题在下面章节报,不要在这里重复

历史 Findings 修复情况

Finding	问题	状态
F1	_destroy_gdr_handle 会静默吞掉 cleanup 失败。	⚠️ 仍存在

📝 PR 规范检查

不符合规范：当前标题 [Cherry-Pick][RL] Reuse GDR checkpoint transfer handle 缺少 release 分支 Cherry-Pick 标题要求的来源 develop PR 号，规范格式为 [Cherry-Pick][Tag] 标题描述(#原PR号)。当前上下文和本地 git 历史没有可验证的来源 PR 号，因此不臆造编号；请补齐真实来源 PR 号后使用该格式。PR 描述已包含 Motivation、Modifications、Usage or Command、Accuracy Tests、Checklist，结构符合 §D2。

总体评价

代码实现和新增单测覆盖了 handle 复用及失败后重建的主要路径；未发现新的阻塞性代码问题。历史 cleanup 异常静默问题仍保留，建议后续至少记录日志以便排障。

[codecov-commenter]

Codecov Report

❌ Patch coverage is 82.60870% with 4 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (release/2.6@ae64731). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
fastdeploy/rl/dynamic_weight_manager.py	82.60%	3 Missing and 1 partial ⚠️

Additional details and impacted files

@@              Coverage Diff               @@
##             release/2.6    #8079   +/-   ##
==============================================
  Coverage               ?   71.60%           
==============================================
  Files                  ?      386           
  Lines                  ?    55790           
  Branches               ?     8764           
==============================================
  Hits                   ?    39951           
  Misses                 ?    13023           
  Partials               ?     2816           

Flag	Coverage Δ
GPU	71.60% <82.60%> (?)

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

☔ View full report in Codecov by Harness.
📢 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.