[Feature] Support set PREEMPTED_TOKEN_ID in GET_SAVE_OUTPUT_V1

[rainyfly]

Motivation

💡 If this PR is a Cherry Pick, the PR title needs to follow the format by adding the [Cherry-Pick] label at the very beginning and appending the original PR ID at the end. For example, [Cherry-Pick][CI] Add check trigger and logic(#5191)

💡 如若此PR是Cherry Pick，PR标题需遵循格式，在最开始加上[Cherry-Pick]标签，以及最后面加上原PR ID，例如[Cherry-Pick][CI] Add check trigger and logic(#5191)

Modifications

Usage or Command

Accuracy Tests

Checklist

• Add at least a tag in the PR title.
  • Tag list: [[FDConfig],[APIServer],[Engine], [Scheduler], [PD Disaggregation], [Executor], [Graph Optimization], [Speculative Decoding], [RL], [Models], [Quantization], [Loader], [OP], [KVCache], [DataProcessor], [BugFix], [Docs], [CI], [Optimization], [Feature], [Benchmark], [Others], [XPU], [HPU], [GCU], [DCU], [Iluvatar], [Metax]]
  • You can add new tags based on the PR content, but the semantics must be clear.
• Format your code, run pre-commit before commit.
• Add unit tests. Please write the reason in this PR if no 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!

[fastdeploy-bot]

🤖 AI Code Review | 2026-04-02 20:02 CST

📋 Review 摘要

PR 概述：在 GET_SAVE_OUTPUT_V1 模式下支持设置 PREEMPTED_TOKEN_ID 标记被抢占的请求
变更范围：worker/gpu_model_runner.py 的 _postprocess 方法
影响面 Tag：Engine Worker

问题

级别	文件	概述
🔴 Bug	gpu_model_runner.py:2407	环境变量名错误，envs.GET_SAVE_OUTPUT_V1 不存在

总体评价

PR 意图清晰，为 V1 输出模式增加了抢占标记支持。但存在一个关键的变量名错误，会导致运行时异常，需要修复后才能合并。

[fastdeploy-bot]

🔴 Bug 变量名错误：envs.GET_SAVE_OUTPUT_V1 不存在，应为 envs.FD_USE_GET_SAVE_OUTPUT_V1

fastdeploy/envs.py 中定义的环境变量是 FD_USE_GET_SAVE_OUTPUT_V1，当前代码使用 envs.GET_SAVE_OUTPUT_V1 会在运行时触发 AttributeError，导致条件分支永远无法进入（Python 会抛出异常）。

其他使用此环境变量的代码（如同文件的 241 行和 2300 行）都使用了正确的 envs.FD_USE_GET_SAVE_OUTPUT_V1。

建议修复：

if envs.FD_USE_GET_SAVE_OUTPUT_V1:

[fastdeploy-bot]

🤖 AI Code Review | 2026-04-03 15:24 CST

📋 Review 摘要

PR 概述：在 FD_USE_GET_SAVE_OUTPUT_V1 模式下，为被抢占的请求设置 PREEMPTED_TOKEN_ID (-9) 以通知 server 端 abort 操作已完成。

变更范围：fastdeploy/worker/gpu_model_runner.py - _postprocess 方法

影响面 Tag：Engine Scheduler

📝 PR 规范检查

PR 描述中 Motivation 和 Modifications 部分未填写具体内容，建议补充说明。

描述模板（可直接复制）：

## Motivation
在使用 GET_SAVE_OUTPUT_V1 模式时，当请求被抢占（preempted）后，没有对应的采样 token。本 PR 通过设置 PREEMPTED_TOKEN_ID (-9) 来通知 server 端 abort 操作已完成，使抢占流程能够正确结束。

## Modifications
- 在 `gpu_model_runner.py` 的 `_postprocess` 方法中，当 `FD_USE_GET_SAVE_OUTPUT_V1` 开启时，检查 `last_preempted_idx`
- 对于被抢占的请求，将其 `sampled_token_ids` 设置为 `PREEMPTED_TOKEN_ID`

问题

未发现阻塞性问题。

总体评价

代码逻辑清晰，与已有的 token_processor.py 中处理 PREEMPTED_TOKEN_ID 的逻辑保持一致。建议补充 PR 描述以便于后续维护和代码审查。

[codecov-commenter]

Codecov Report

❌ Patch coverage is 66.66667% with 1 line in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@938e7dd). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
fastdeploy/worker/gpu_model_runner.py	66.66%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #7159   +/-   ##
==========================================
  Coverage           ?   73.86%           
==========================================
  Files              ?      376           
  Lines              ?    52888           
  Branches           ?     8250           
==========================================
  Hits               ?    39064           
  Misses             ?    11095           
  Partials           ?     2729           

Flag	Coverage Δ
GPU	73.86% <66.66%> (?)

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.