feat(sources): Wayback pass, automatic trimming, detention rule, ten cases re-sourced

Second and third rounds on the dossier pipeline, reviewed locally before commit.

Retrieval
- wayback_pass.py: retries every BLOCKED/DEAD/THIN/ERROR linked source through
  the Wayback Machine; blocks carry the snapshot date, archive URL and the live
  verdict. 143 slots retried: 93 fetched, 37 no snapshot, 12 script shells, 1 PDF.
- fetch_sources.py: fetch_wayback() helper.
- add_extra_sources.py: files staged extra_* finds (web search, not on the DB
  record) into the store with a provenance note.
- Cases re-sourced by web search: 248 (Oslo: Document.no, Avisa Oslo, NRK; the
  linked Le Parisien piece is case 242 and is marked OFF-CASE), 258 Kharkiv,
  260 Singapore, 358 Bangkok, 365 Las Vegas, 388 Verneuil-sur-Seine, 390 Zoersel,
  430 Homestead. 341 of 364 cases now hold a fetched article; 11, 66 and 399
  have no public text source (podcast, police video, direct victim report).

Trimming
- auto_trim.py: case-anchored furniture cut for UNTRIMMED blocks. Finds the body
  run that mentions the case, merges across subheadings and short furniture
  gaps, drops teasers, share bars, date/URL/caption lines and subscription
  pitches; refuses pages with no record term or almost no body. 302 blocks
  trimmed; 16 left on auto-trim-review.md (7 OFF-CASE suspects). Every cut is
  labelled AUTO-TRIMMED, UNREVIEWED in the dossier; the full extract stays in
  staging/. Decisions with anchors in trim-decisions-auto.json.
- build_cases.py / verify_all.py: banners for auto-trimmed and archived blocks,
  staged-file check extended to Wayback blocks, flag legend under the record.

Detention rule
- README "Definitions": the DB field `kidnappings` is the DETENTION violence
  type (victim, guard, staff or relative held to force submission or execute
  the theft), distinct from the Kidnapping scenario (taken away and held).
- detention-flag-review.md / detention-flag-corrections.json: 40 records
  reviewed with evidence; 30 set-to-1 proposals accepted by the owner on
  2026-09-06 (listed in corrections-approved.md), 10 still open.
- apply_detention_wording.py: "Violence Used" in the 70 reviewed summaries now
  names detention explicitly (62 of 70 labelled), supported by the summary's
  own text; idempotent; supersedes the batch scripts' wording.

Worklist and docs
- sofia-worklist.md: 248 decisions, detention rule item replacing the old
  "no abduction" item, fresh-search section for the textless cases.
- README-START-HERE.md: progress notes, run commands, next steps.
- Bug fixed in passing: Wayback blocks stored in-memory text with carriage
  returns; now stored as read back from disk.

verify_all.py PASSES (2321 checks).

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GZZENdTLzNGsbNy4DyF1yt

merge-output/wayback-progress.json

@@ -0,0 +1,145 @@
+{
+  "451:url": "THIN",
+  "449:url": "FETCHED",
+  "448:url": "THIN",
+  "446:url": "THIN",
+  "445:url": "THIN",
+  "444:url": "THIN",
+  "442:url_2": "FETCHED",
+  "441:url": "FETCHED",
+  "440:url": "FETCHED",
+  "440:url_2": "FETCHED",
+  "439:url": "FETCHED",
+  "439:url_2": "FETCHED",
+  "438:url_2": "FETCHED",
+  "438:url_3": "FETCHED",
+  "437:url": "FETCHED",
+  "436:url_3": "NO SNAPSHOT",
+  "431:url": "NO SNAPSHOT",
+  "431:url_2": "NO SNAPSHOT",
+  "430:url": "NO SNAPSHOT",
+  "427:url_2": "THIN",
+  "426:url": "FETCHED",
+  "426:url_2": "NO SNAPSHOT",
+  "424:url_2": "FETCHED",
+  "423:url_2": "FETCHED",
+  "422:url_2": "FETCHED",
+  "417:url_2": "FETCHED",
+  "416:url": "THIN",
+  "414:url": "FETCHED",
+  "412:url": "NO SNAPSHOT",
+  "412:url_4": "NO SNAPSHOT",
+  "412:url_5": "NO SNAPSHOT",
+  "411:url_2": "FETCHED",
+  "410:url": "THIN",
+  "408:url": "NO SNAPSHOT",
+  "407:url_3": "FETCHED",
+  "407:url_4": "FETCHED",
+  "405:url_2": "NO SNAPSHOT",
+  "402:url_3": "NO SNAPSHOT",
+  "395:url": "FETCHED",
+  "393:url_2": "FETCHED",
+  "391:url": "FETCHED",
+  "390:url": "NO SNAPSHOT",
+  "390:url_2": "NO SNAPSHOT",
+  "389:url_2": "FETCHED",
+  "388:url": "NO SNAPSHOT",
+  "387:url": "FETCHED",
+  "384:url_2": "NO SNAPSHOT",
+  "383:url_2": "NO SNAPSHOT",
+  "382:url": "FETCHED",
+  "382:url_2": "THIN",
+  "382:url_3": "NO SNAPSHOT",
+  "382:url_4": "NO SNAPSHOT",
+  "381:url": "FETCHED",
+  "379:url_3": "FETCHED",
+  "372:url": "FETCHED",
+  "370:url": "THIN",
+  "369:url": "FETCHED",
+  "367:url": "FETCHED",
+  "365:url": "NO SNAPSHOT",
+  "363:url_4": "NO SNAPSHOT",
+  "361:url": "FETCHED",
+  "360:url_2": "FETCHED",
+  "258:url": "NO SNAPSHOT",
+  "45:url": "FETCHED",
+  "355:url": "FETCHED",
+  "354:url": "FETCHED",
+  "349:url": "FETCHED",
+  "323:url": "FETCHED",
+  "322:url": "FETCHED",
+  "318:url": "NO SNAPSHOT",
+  "307:url": "FETCHED",
+  "293:url": "FETCHED",
+  "290:url": "FETCHED",
+  "288:url": "FETCHED",
+  "285:url": "FETCHED",
+  "271:url": "FETCHED",
+  "263:url": "FETCHED",
+  "256:url": "FETCHED",
+  "255:url_2": "FETCHED",
+  "255:url_3": "FETCHED",
+  "253:url": "NO SNAPSHOT",
+  "248:url": "FETCHED",
+  "247:url": "FETCHED",
+  "237:url": "NO SNAPSHOT",
+  "229:url": "FETCHED",
+  "228:url_2": "FETCHED",
+  "223:url": "FETCHED",
+  "222:url": "FETCHED",
+  "219:url": "FETCHED",
+  "218:url": "FETCHED",
+  "214:url": "NO SNAPSHOT",
+  "213:url": "FETCHED",
+  "212:url": "FETCHED",
+  "210:url": "FETCHED",
+  "209:url_4": "FETCHED",
+  "207:url": "NO SNAPSHOT",
+  "199:url": "FETCHED",
+  "197:url": "NO SNAPSHOT",
+  "195:url": "FETCHED",
+  "194:url": "FETCHED",
+  "188:url_2": "FETCHED",
+  "188:url_3": "FETCHED",
+  "183:url": "FETCHED",
+  "179:url": "NO SNAPSHOT",
+  "176:url": "NO SNAPSHOT",
+  "168:url": "FETCHED",
+  "164:url": "FETCHED",
+  "162:url": "FETCHED",
+  "160:url": "FETCHED",
+  "157:url": "NO SNAPSHOT",
+  "152:url": "FETCHED",
+  "151:url": "FETCHED",
+  "145:url": "FETCHED",
+  "141:url": "FETCHED",
+  "138:url": "NO SNAPSHOT",
+  "116:url": "FETCHED",
+  "113:url": "FETCHED",
+  "111:url": "FETCHED",
+  "110:url": "FETCHED",
+  "105:url": "FETCHED",
+  "101:url": "FETCHED",
+  "91:url": "FETCHED",
+  "80:url": "NO SNAPSHOT",
+  "79:url": "THIN",
+  "71:url": "FETCHED",
+  "69:url": "FETCHED",
+  "67:url": "NO SNAPSHOT",
+  "62:url": "FETCHED",
+  "60:url": "FETCHED",
+  "57:url": "NO SNAPSHOT",
+  "56:url": "FETCHED",
+  "54:url": "FETCHED",
+  "44:url": "FETCHED",
+  "41:url": "FETCHED",
+  "40:url": "FETCHED",
+  "33:url": "FETCHED",
+  "32:url": "NO SNAPSHOT",
+  "30:url": "NO SNAPSHOT",
+  "21:url": "NO SNAPSHOT",
+  "18:url": "PDF",
+  "16:url": "FETCHED",
+  "9:url": "THIN",
+  "6:url": "FETCHED"
+}