stop processing history

explictly del df to fix oom

fix viz

cleanup metrics and fix repo

rename OpenDevin to OpenHands

fix model_name in updated metadat

support loading report with new format

change test_result to bool

fix fine-grained report; support visualization while running

support visualization of new swebench-eval

Merge commit 'f6d9f43457bdadd36685181efda2fd45e813a02c'

visualize swe-bench-lite & fix stuck in look

add cost info when exists

show errrors

add absolute number of solved

add benchmark code

support multi-page