[BugFix] fix speculate_limit_thinking_content_length

[yuanlehome]

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

This PR removes an unused seq_lens_decoder parameter from the speculate_limit_thinking_content_length family of functions, which are used to limit thinking content length during speculative decoding.

Key Changes:

• Removed seq_lens_decoder parameter from v1 and v2 CUDA kernel functions and their wrappers
• Updated all test cases to remove the unused parameter
• Removed redundant sequence length adjustment logic (now only step_idx is adjusted)

Critical Issue Found: The function call in post_process_specualate is missing required parameters after the removal.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

File	Description
tests/operators/test_speculate_limit_thinking_content_length.py	Removed seq_lens_decoder parameter from all test cases for both v1 and v2 functions; updated comments to reflect removal
fastdeploy/model_executor/pre_and_post_process.py	Removed seq_lens_decoder parameter from function signatures and internal calls; bug introduced in call at line 444
custom_ops/gpu_ops/speculate_decoding/speculate_limit_thinking_content_length_v1.cu	Removed seq_lens_decoder from kernel, wrapper function, and operator input declaration; removed redundant adjustment logic
custom_ops/gpu_ops/speculate_decoding/speculate_limit_thinking_content_length_v2.cu	Removed seq_lens_decoder from kernel, wrapper function, and operator input declaration; removed redundant adjustment logic

Note on PR Description: The PR description is incomplete. The "Motivation" and "Modifications" sections are empty. Please add:

1. Motivation: Explain that seq_lens_decoder was redundant because it was being updated in the same way as step_idx, and the actual sequence length update happens in the speculate_update function that is called after this function.
2. Modifications: Detail that this PR removes the unused seq_lens_decoder parameter from the speculate_limit_thinking_content_length_v1 and v2 functions across CUDA kernels, Python wrappers, and test files.

[Copilot]

Missing required parameters in function call. The function speculate_limit_thinking_content_length requires stop_flags and eos_token_ids parameters (as seen in the function signature on line 147-148), but they are not provided in this call. After removing seq_lens_decoder, you should add:

• stop_flags=share_inputs["stop_flags"], after line 450
• eos_token_ids=share_inputs["eos_token_ids"], after stop_flags (only needed for the "" strategy, but the function signature requires it)

This will cause a runtime error when this function is called.

Suggested change

	line_break_id=line_break_id,
	line_break_id=line_break_id,
	stop_flags=share_inputs["stop_flags"],
	eos_token_ids=share_inputs["eos_token_ids"],

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (develop@c9b47f9). Learn more about missing BASE report.

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #5590   +/-   ##
==========================================
  Coverage           ?   61.93%           
==========================================
  Files              ?      328           
  Lines              ?    41134           
  Branches           ?     6270           
==========================================
  Hits               ?    25476           
  Misses             ?    13743           
  Partials           ?     1915           

Flag	Coverage Δ
GPU	61.93% <ø> (?)

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.