real_006: distribution-shift test on rows 10-59 of fill_events

Methodology fix: gen_real_queries.go gains -offset N flag. Every prior
real_NNN test sourced queries from rows 0-9 of fill_events.parquet
(default -limit 10), so the substrate's published "8/10 cold-pass top-1
= judge-best" was measured on a memorized slice, not held-out data.

real_006 samples 50 fresh rows (offset 10, never seen by the workers
or ethereal_workers corpora). Same harness, same local qwen2.5:latest
judge, same K=10. ~14 min wall total. Local-only, no cloud calls.

Headline findings:

- Cold-pass top-1 = judge-best (rank match): 41/50 (82%) vs real_001's
  8/10 (80%) — substrate generalizes at rank level.
- Strict (rating ≥ 2): 34/50 (68%) — 12-point drop from real_001's
  80%. ~7 of 41 "no-discovery" queries had cold top-1 the judge rated
  1; the corpus has gaps for some role-city combos in the v3 slice.
- Verbatim lift: 9/9 discoveries → warm top-1 (clean, matches real_001 2/2)
- Paraphrase recovery: 6/9 → top-1, 9/9 any-rank
- Quality regressed: 3/50 — Q43 is the structural one

Q43 (Packer at Midway Distribution / Chicago IL) regressed from
rating 5 to rating 2 on warm pass with `warm_boosted_count=0` and
`playbook_recorded=false`. Q18 (Shipping Clerks at the same client+city)
recorded a playbook entry. The regression suggests Q18's recording
leaked into Q43 via the warm-pass playbook corpus retrieval surface
even though the role gate from real_002 should have blocked it.
Three possible paths: extractor failed on one query, gate fires on
boost path but not Shape B inject, or cosine drift puts the recorded
worker close enough to Q43's embedding that warm-pass retrieval picks
it up directly. Diagnosis is the next move.

Three same-(client, city) clusters tested:
- Heritage Foods Gary IN × 3 distinct roles: clean, distinct workers
- Riverfront Steel Columbus OH × 4: cosine-level confusion (Q9/Q25
  surface same worker w-281 for Assemblers vs Quality Techs at cold-
  pass), but no playbook bleed
- Midway Distribution Chicago IL × 3: Q43 regression as above

What this confirms: substrate works on the fresh distribution at the
rank level, verbatim lift is real, paraphrase recovery is real.

What this falsifies: real_002's role-gate fix is not structurally
airtight. The bleed pattern can still fire under conditions the
prior tests didn't reach.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

scripts/cutover/gen_real_queries.go

@@ -29,6 +29,7 @@ import (
 func main() {
 	src := flag.String("src", "/home/profit/lakehouse/data/datasets/fill_events.parquet", "fill_events parquet path")
 	limit := flag.Int("limit", 10, "number of source rows to read")
+	offset := flag.Int("offset", 0, "skip the first N rows (lets reality tests sample beyond the memorized real_001 slice)")
 	styles := flag.String("styles", "need", "comma-separated styles to emit per row (need|client_first|looking|shorthand|all)")
 	flag.Parse()
 
@@ -58,16 +59,24 @@ func main() {
 	at := tbl.Column(10).Data().Chunk(0)
 	deadline := tbl.Column(12).Data().Chunk(0)
 
-	n := int(tbl.NumRows())
-	if *limit < n {
-		n = *limit
+	totalRows := int(tbl.NumRows())
+	start := *offset
+	if start < 0 {
+		start = 0
+	}
+	if start > totalRows {
+		start = totalRows
+	}
+	end := start + *limit
+	if end > totalRows {
+		end = totalRows
 	}
 
 	stylesList := parseStyles(*styles)
 
 	fmt.Println("# Real-shape coordinator queries — generated from fill_events.parquet")
 	fmt.Println("# (real-shape demand data; queries built mechanically from event rows).")
-	fmt.Printf("# Source: %s (%d rows total, %d emitted, styles=%v)\n", *src, tbl.NumRows(), n, stylesList)
+	fmt.Printf("# Source: %s (%d rows total, rows [%d,%d) emitted, styles=%v)\n", *src, totalRows, start, end, stylesList)
 	fmt.Println("#")
 	fmt.Println("# Styles:")
 	fmt.Println("#   need:         'Need N {role}{s} in {city} {state} starting at {at} for {client}'")
@@ -80,7 +89,7 @@ func main() {
 	fmt.Println("# substrate's bleed behavior when the role gate is silently disabled.")
 	fmt.Println()
 
-	for i := 0; i < n; i++ {
+	for i := start; i < end; i++ {
 		ev := event{
 			client: client.ValueStr(i),
 			city:   city.ValueStr(i),