Revert "[Feature] Support computing entropy with fastdeploy runner"

[rain7996]

Reverts #7954
Revert以恢复对GLM RL精度的影响

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[PaddlePaddle-bot]

🤖 Paddle-CI-Agent | pr_review | 2026-06-03 20:17:44

📋 Review 摘要

PR 概述：Revert PR #7954，移除 FD runner 的 entropy 计算路径，以恢复 GLM RL 精度。
变更范围：model_executor/entropy_utils.py、model_executor/pre_and_post_process.py、worker/gpu_model_runner.py
影响面 Tag：[BugFix] [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。

📝 PR 规范检查

标题缺少 FastDeploy 官方 Tag，且 PR body 缺少所有必填 section（## Motivation、## Modifications、## Usage or Command、## Accuracy Tests、## Checklist）。

标题建议（可直接复制）：

• [BugFix] Revert entropy computing with fastdeploy runner to fix GLM RL accuracy

PR 描述建议（点击展开，可直接复制）

## Motivation
PR #7954 引入了通过 `EB5_ENABLE_FD_RUNNER` 环境变量控制的 FD runner entropy 计算路径，该路径导致 GLM RL 精度下降，需要 revert。

## Modifications
- `fastdeploy/model_executor/entropy_utils.py`：移除 `calculate_logits_entropy_fd`、`speculate_calculate_logits_entropy_fd`、`flush_entropy_on_stop` 函数及 `_log_entropy` helper；恢复 `calculate_logits_entropy` 和 `speculate_calculate_logits_entropy` 原始实现
- `fastdeploy/model_executor/pre_and_post_process.py`：移除 `flush_entropy_on_stop` 的 import 和调用
- `fastdeploy/worker/gpu_model_runner.py`：恢复 `_dummy_prefill_inputs` 中 `seq_lens_this_time` 不做 `[:batch_size]` 截断的赋值
- `tests/model_executor/test_entropy_utils_fd_runner_mtp.py`：删除随 #7954 新增的 FD runner 路径测试

## Usage or Command
N/A

## Accuracy Tests
N/A（Revert 以恢复 GLM RL 精度，精度对比数据由原 PR 作者提供）

## Checklist

- [x] 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.
- [x] 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.

总体评价

本 Revert 实现正确，所有功能代码改动与 #7954 完全对称，gpu_model_runner.py 中去掉 [:batch_size] 截断与移除 FD runner entropy 路径一致，无正确性问题。PR 规范（标题 Tag 缺失、描述模板未填写）需修复。