Support penalty in overlap mode; return logprob with chunked prefill; improve benchmark scripts (#3988)

Co-authored-by: SangBin Cho <rkooo567@gmail.com>
Co-authored-by: dhou-xai <dhou@x.ai>
Co-authored-by: Hanming Lu <hanming_lu@berkeley.edu>

python/pyproject.toml

@@ -96,7 +96,10 @@ dev_cpu = ["sglang[all_cpu]", "sglang[test]"]
 "Bug Tracker" = "https://github.com/sgl-project/sglang/issues"
 
 [tool.setuptools.package-data]
-"sglang" = ["srt/layers/moe/fused_moe_triton/configs/*.json", "srt/layers/quantization/configs/*.json"]
+"sglang" = [
+    "srt/layers/moe/fused_moe_triton/configs/*.json",
+    "srt/layers/quantization/configs/*.json",
+]
 
 [tool.setuptools.packages.find]
 exclude = [