[Feature] Support mtp overlap schedule

[Sunny-bot1]

Motivation

支持MTP场景下开启overlap schedule

Modifications

位置	算子	DtoH操作	解决方法
gpu_model_runner.py 前处理	unified_update_model_status	not_need_stop	废弃
mtp.py 前处理	draft_model_preprocess	not_need_stop	废弃
	eagle_get_hidden_states	output_token_num_cpu	pre_alloc (WIP)
	_propose_cuda（token_num_cpu）	self.model_inputs["seq_lens_this_time"].numpy().sum().item()	异步拷贝，复用上一轮的token_num_cpu（seq_lens_this_time.sum）
mtp.py 后处理	draft_model_update	not_need_stop	废弃
gpu_model_runner.py 后处理	speculate_schedule_cache	not_need_stop	废弃
	save_output	accept_tokens_cpu, accept_num_cpu, seq_lens_decoder_cpu, prompt_lens_cpu	异步拷贝，延迟传输
worker_process.py	tp_barrier	all_reduce	开启overlap schedule后自动使用CPU barrier
cudagraph_piecewise_backend.py	call（num_running_requests）	num_running_requests = int((seq_lens_this_time.flatten() > 0).sum().item())	异步拷贝，复用上一轮的num_running_requests（(seq_lens_this_time>0).sum）

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!

[codecov-commenter]

Codecov Report

❌ Patch coverage is 77.06422% with 25 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@cffa8c2). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
fastdeploy/worker/input_batch.py	47.82%	12 Missing ⚠️
fastdeploy/model_executor/pre_and_post_process.py	54.54%	4 Missing and 1 partial ⚠️
.../graph_optimization/cudagraph_piecewise_backend.py	40.00%	2 Missing and 1 partial ⚠️
fastdeploy/spec_decode/mtp.py	91.89%	2 Missing and 1 partial ⚠️
fastdeploy/worker/gpu_model_runner.py	93.75%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #6785   +/-   ##
==========================================
  Coverage           ?   72.28%           
==========================================
  Files              ?      395           
  Lines              ?    54772           
  Branches           ?     8617           
==========================================
  Hits               ?    39591           
  Misses             ?    12355           
  Partials           ?     2826           

Flag	Coverage Δ
GPU	72.28% <77.06%> (?)

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.