[BugFix] [PD Disaggregation] Fix schedule error in splitwise deployment

[juncaipeng]

Motivation

修复 PD 分离下调度问题，从 #5027 进行拆分。

Modifications

• 调度循环中注意释放cache block，避免block 不够
• local_scheduler 拉取至少一条请求，避免一直拉不到请求

Usage or Command

没有变化

Accuracy Tests

单侧覆盖

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 fixes scheduling issues in PD disaggregation (splitwise deployment) by addressing two key problems: block resource management and request pulling logic.

• Ensures cache blocks are freed at the start of each scheduling cycle to prevent resource exhaustion
• Guarantees that the local scheduler pulls at least one request before applying token budget constraints

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
fastdeploy/scheduler/local_scheduler.py	Modified request pulling logic to ensure at least one request is retrieved before breaking on token budget limit
fastdeploy/engine/sched/resource_manager_v1.py	Added block table cleanup at the beginning of the schedule loop to free cache blocks when needed