workflow: OCR with alternatives, n-best decoding context classifier, character whitelisting re-OCR, no threshold

workflow: add cropping, remove fixed DPI, remove numeric-only OCR constraint, output OCR character choices

avoid system resdir as long as it's defined as VOLUME; ensure minimal tessdata files are in new default location

first functional version; fixed a-priori 1:1:1 model:port:server dependency

model files kept separate (only copied into worktree to build Docker image)