rename package doc_agent -> docfield, distribution -> docfield-extract

docs: report tuning and held-out separately, with the held-out result as headline

eval: tick F12, correct FC-1's scope, document FC-3

eval: add --retry-errors to re-predict only the entries that failed

logging: keep extracted field values out of log records

space: configure logging so the per-document record actually reaches the log

docs: update README

fix: shorten readme frontmatter short_description to below 60 chars

web: trim docstrings and drop the copy-preference test

web: explain the decision in the output, in plain language

docs: drop remaining "agent" framing from shipped docstrings and backlog

validation: monetary tolerance scales with rounded terms, not with value

eval: correct FC-1's corpus claim, log FC-2, note cent-exact severity

eval: document failure case FC-1 and log its open questions

eval: report documents that never reached the model, grouped by cause

validation: S2 skips an absent currency instead of failing it

eval: add tuning/held-out split selection to the score command

eval: score under current rules with --revalidate, and detect cache drift

docs: rename to DocField Extract, reframe as key-value extraction pipeline

docs: note specs are frozen inputs, code is authoritative

docs: update readme and add MIT license

docs: rewrite README as portfolio hero page

docs: track PROGRESS_FUTURE.md backlog

docs: correct stale money-comparison docstring in eval.normalize

chore: tick T11 -- Space live, both modalities verified end-to-end

docs: add live Space URL to README

fix(space): pin pydantic==2.12.5 to resolve against gradio mcp extra

phase 6: curate Space requirements + fix src import (T11)

phase 6: Hugging Face Space deploy scaffolding (T11)

phase 5: set CONFIDENCE_THRESHOLD=0.50, record SROIE results (T10)

phase 5: evaluation harness (SROIE)

feat(web): Gradio stateless web demo (T9)

fix(lint): clear 5 ruff F401/F841 errors found by verifier

feat(watcher): folder watcher + batch runner (T8)

feat(store): SQLite persistence + CSV export (T7)

chore: tick T5 (acquire consolidation done incrementally in T2/T3)

feat(docling): Docling parser + native-PDF acquire path (T3)

feat(gemini): Gemini backend + vision-direct image acquire (T2)

chore: tick T1 checkbox in PROGRESS_TOMORROW.md

feat(deps): add google-genai; wire Gemini config (T1)

chore: overnight loop harness + ignore run logs

tick N-FINAL: full suite + lint green

tick N9: content-hash idempotency helper

phase 3.2: content-hash idempotency helper

tick N8: core pipeline orchestration

phase 3.1: core pipeline orchestration (stub-backed smoke test)

tick N7: backend interface + offline stub

phase 2.4: backend interface + factory + offline stub backend

tick N6: modality detection

phase 2.1: modality detection