[sgl] improve accuracy of additional page requirement during spec decode

[2022tgoel]

Motivation

The new_tokens_required_next_decode calculation was very conservative in calculating how many pages a batch will consume. I would like to replace it with a more realistic estimate that mimics the logic in eagle_info_v2.py

Modifications

Accuracy Tests

Speed Tests and Profiling

Checklist

• Format your code according to the Format code with pre-commit.
• Add unit tests according to the Run and add unit tests.
• Update documentation according to Write documentations.
• Provide accuracy and speed benchmark results according to Test the accuracy and Benchmark the speed.
• Follow the SGLang code style guidance.

Review and Merge Process

1. Ping Merge Oncalls to start the process. See the PR Merge Process.
2. Get approvals from CODEOWNERS and other reviewers.
3. Trigger CI tests with comments or contact authorized users to do so.
   • Common commands include /tag-and-rerun-ci, /tag-run-ci-label, /rerun-failed-ci
4. After green CI and required approvals, ask Merge Oncalls or people with Write permission to merge the PR.

[gemini-code-assist]

Warning

You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again!

[ispobock]

@2022tgoel could you fix the lint first?

[Qiaolin-Yu]

I feel like it's correct. But also cc @hnyls2002 for another check in case I might be missing some context.

[hnyls2002]

/tag-and-rerun-ci

[hnyls2002]

/rerun-test test_eagle_infer_a.py test_eagle_infer_b.py test_eagle_infer_beta.py test_eagle3_basic.py test_specv2_kvcache_offloading.py test_swa_radix_cache_kl.py test_swa_unittest.py test_eagle_dp_attention.py

[github-actions]

✅ 1-gpu-h100: View workflow run

cd test/ && python3 registered/spec/eagle/test_eagle_infer_a.py

✅ 1-gpu-h100: View workflow run

cd test/ && python3 registered/spec/eagle/test_eagle_infer_b.py

✅ 1-gpu-5090: View workflow run

cd test/ && python3 registered/spec/eagle/test_eagle_infer_beta.py

✅ 1-gpu-5090: View workflow run

cd test/ && python3 registered/spec/eagle/test_eagle3_basic.py

✅ 1-gpu-5090: View workflow run

cd test/ && python3 registered/disaggregation/test_specv2_kvcache_offloading.py

✅ 1-gpu-h100: View workflow run

cd test/ && python3 registered/radix_cache/test_swa_radix_cache_kl.py

✅ 1-gpu-h100: View workflow run

cd test/ && python3 registered/unit/mem_cache/test_swa_unittest.py

✅ 4-gpu-h100: View workflow run

cd test/ && python3 registered/spec/eagle/test_eagle_dp_attention.py