[Feature]Add a switch for logprobs/prompt_logprobs token decoding.

[qwes5s5]

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)

It has been observed during current project usage that when logprobs or prompt_logprobs take a particularly large value, the process of converting token IDs to tokens consumes a significant amount of time. In some scenarios, the decoded tokens are not frequently used. Therefore, a switch needs to be added to control whether the decoding from token ID to token is enabled for logprobs and prompt_logprobs on a specific request.

Modifications

Add the control parameter include_logprobs_decode_token to the online link request, which is enabled by default.

• For the v1/chat/completions interface:
  • When include_logprobs_decode_token is enabled, the results for logprobs and prompt_logprobs are output in the normal format.
  • When include_logprobs_decode_token is disabled, the token and bytes fields in the logprobs result are empty, and the decoded_tokens field in prompt_logprobs is empty.
• For the v1/completions interface:
  • When include_logprobs_decode_token is enabled, the results for logprobs and prompt_logprobs are output in the normal format.
  • When include_logprobs_decode_token is disabled, the logprobs result is output in the normal format, and the decoded_tokens field in prompt_logprobs is empty.

Usage or Command

v1/chat/completions：

curl --location 'http://0:0:0:0:8180/v1/completions' \
--header 'Content-Type: application/json' \
--data '{
  "prompt": "请生成一篇关于理想的作文,不超过100字",
  "logprobs":3,
  "prompt_logprobs":3,
   "include_logprobs_decode_token":false
}
'

v1/completions：

curl --location 'http://0:0:0:0:8180/v1/chat/completions' \
--header 'Content-Type: application/json' \
--data '{
    "messages": [
        {
            "role": "system",
            "content": "I'\''m a helpful AI assistant."
        },
        {
            "role": "user",
            "content": "give me three letter randdomly, just tell me letter without anything else"
        }
    ],
    "logprobs":"true",
    "top_logprobs":3,
    "prompt_logprobs":3,
    "include_logprobs_decode_token":false
}'

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!

[codecov-commenter]

Codecov Report

❌ Patch coverage is 59.25926% with 11 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@3066a0c). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
fastdeploy/entrypoints/openai/serving_chat.py	57.14%	6 Missing and 3 partials ⚠️
...astdeploy/entrypoints/openai/serving_completion.py	50.00%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #5463   +/-   ##
==========================================
  Coverage           ?   60.31%           
==========================================
  Files              ?      329           
  Lines              ?    41067           
  Branches           ?     6247           
==========================================
  Hits               ?    24771           
  Misses             ?    14410           
  Partials           ?     1886           

Flag	Coverage Δ
GPU	60.31% <59.25%> (?)

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.