feat: CLI local mode, slash commands, interrupt support; remove lmnr; frontend fixes

fix: properly close SDK message on error, show tool errorText

Revert "fix: show errorText for failed tools, bump eval max_iterations to 300"

fix: show errorText for failed tools, bump eval max_iterations to 300

functioning frontend and docker

generated, filled in and verfied 250 eval questions

intermediate commit until i let amp loose

eval readme update

gpt 5 nano judge

fixing tracing

rename

rename

adding claude code + mcp

leaderboard and results

link fix

updated eval

dataset creation script

adding readme

adding hf datasets i/o

eval script done

eval runs

modified eval prompt

thinking if we want eval or not