chore: Release paperless-llm-workflows version 0.3.3

update CHANGELOG for pending release

update default model in container build to best performing model from benchmarking

cleanup unused imports

upgared llama-cpp-rs binding to lastet version

Merge branch 'feature/bench-eval-scripts'

add README describing how to use the evaluation scripts

typst script to generate result table

plot benchmark results python script

implement resume multi-benchmark functionality

save benchmark subprocess exits and record error source

update llama-cpp-rs dependency

endless event log

Merge branch 'feature/model-benchmarking'

Add benchmarking documentation to README