[Optimization]Optimize CPU utilization

[luukunn]

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!

[Copilot]

Pull request overview

该 PR 旨在通过减少解码阶段的临时对象创建与重复反射判断，优化服务端在流式/非流式输出处理时的 CPU 开销，涉及输入处理器的增量解码与 OpenAI 响应处理流程。

Changes:

• 在 DataProcessor/Ernie4_5Processor.ids2tokens 中改为对历史 token 列表就地 extend，避免每步生成 previous + token_id 的 O(n) 临时 list。
• 在 ChatResponseProcessor 中缓存 process_response_dict 是否为协程函数，避免在循环中重复 inspect.iscoroutinefunction(...)。
• 对若干内部状态访问进行局部变量缓存（status = self.decode_status[task_id]）以减少字典/索引访问开销。

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
fastdeploy/input/text_processor.py	增量解码路径就地累积 token，减少临时 list 分配
fastdeploy/input/ernie4_5_processor.py	同步增量解码路径同样改为就地累积 token
fastdeploy/entrypoints/openai/response_processors.py	缓存协程判断，减少循环内反射开销（同时暴露出一个需要修复的同步分支问题）

[Copilot]

这里改为使用 self._is_async_processor 后，同一代码块下方的同步分支（对应 else）仍然调用 process_response_dict 时传入了外层的 request_output/stream 变量，而不是当前 part["request_output"]（且应固定 stream=False），会导致非流式 multipart(text+image+text) 场景文本解码使用错误输入。建议把同步分支也改成对每个 part 调用 process_response_dict(response_dict=part["request_output"], stream=False, ...) 以与异步路径一致。

[codecov-commenter]

Codecov Report

❌ Patch coverage is 68.00000% with 8 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@2b10ebc). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
fastdeploy/input/text_processor.py	61.53%	5 Missing ⚠️
...stdeploy/entrypoints/openai/response_processors.py	25.00%	0 Missing and 3 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #6950   +/-   ##
==========================================
  Coverage           ?   73.73%           
==========================================
  Files              ?      399           
  Lines              ?    55624           
  Branches           ?     8766           
==========================================
  Hits               ?    41017           
  Misses             ?    11707           
  Partials           ?     2900           

Flag	Coverage Δ
GPU	73.73% <68.00%> (?)

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.