[Quantization] Support w4afp8 moe weight offline permute & load

[Sunny-bot1]

Motivation

• 为了提升模型加载速度以及适配EPLB，支持离线量化重排，加载权重后无需其他操作
• 支持dense在线量化、moe离线量化

Modifications

is_quantized：全局控制是否走离线量化逻辑

is_moe_quantized：控制moe的权重是否走离线量化逻辑

moe_dynamic_quant：控制moe的激活是否走在线量化逻辑

is_permuted：控制moe的权重是否离线重排

Usage or Command

w4afp8的量化组合：

dense block_wise_fp8权重	moe w4afp8激活	moe w4afp8权重	quantization_config
在线	在线	在线	"quantization_config":{ "dense_quant_type":"block_wise_fp8", "moe_quant_type":"w4afp8", "quantization":"mix_quant","hadamard_block_size": 512},
在线	在线	离线不重排	"quantization_config":{"dense_quant_type":"block_wise_fp8", "moe_quant_type":"w4afp8", "is_moe_quantized":true, "moe_dynamic_quant": true, "quantization":"mix_quant", "hadamard_block_size": 512, "is_permuted":false}
在线	在线	离线重排	"quantization_config":{"dense_quant_type":"block_wise_fp8", "moe_quant_type":"w4afp8", "is_moe_quantized":true, "moe_dynamic_quant": true,"quantization":"mix_quant","hadamard_block_size": 512, "is_permuted":true }
在线	离线	离线重排	"quantization_config":{"dense_quant_type":"block_wise_fp8", "moe_quant_type":"w4afp8", "is_moe_quantized":true, "quantization":"mix_quant", "hadamard_block_size": 512, "is_permuted":true}
离线	离线	离线重排	"is_quantized": true, "quantization_config":{"dense_quant_type":"block_wise_fp8", "moe_quant_type":"w4afp8", "quantization":"mix_quant","hadamard_block_size": 512, "is_permuted":true}

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 13.63636% with 19 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@4f830aa). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
...l_executor/layers/moe/fused_moe_cutlass_backend.py	0.00%	19 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #5613   +/-   ##
==========================================
  Coverage           ?   64.57%           
==========================================
  Files              ?      329           
  Lines              ?    41780           
  Branches           ?     6397           
==========================================
  Hits               ?    26978           
  Misses             ?    12755           
  Partials           ?     2047           

Flag	Coverage Δ
GPU	64.57% <13.63%> (?)

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.

[yangjianfengo1]

LGTM