[RL][CI] Support Async R3 And Add Accuracy Test

[gongshaotian]

Motivation

1. Preformance: Support Async put the routing indices of one query when the self-regression has ended.
2. Accuracy CI: Add Accuracy CI for the EB45 and GLM45 Air Models.
3. Compatibility: Support the implementation of R3 + MTP. Don't record the routing indices generated by the Moe layer of draft model.
4. Add p2pstore to requirements.txt.

TODO:

• Support R3 + Prefix Cache
• Support full async put

Model Support List:

• EB45
• EB5
• GLM
• EB45 vl

Modifications

• Improve RoutingReplayManager Preformance

Usage or Command

Pass

Accuracy Tests

• CI Case：
  • Add Accuracy CI for EB45 Lite
  • Modify test case of GLM45 Air
    • Batch Size: 32 -> 1
    • Quantization: fp8 -> bf 16

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 70.00000% with 6 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@6825903). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
...model_executor/layers/moe/routing_indices_cache.py	72.22%	5 Missing ⚠️
fastdeploy/model_executor/layers/moe/moe.py	50.00%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #5937   +/-   ##
==========================================
  Coverage           ?   66.91%           
==========================================
  Files              ?      348           
  Lines              ?    44662           
  Branches           ?     6867           
==========================================
  Hits               ?    29884           
  Misses             ?    12584           
  Partials           ?     2194           

Flag	Coverage Δ
GPU	66.91% <70.00%> (?)

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.