tools: distinguish "captured in time" from "still alive"

analyze_media_age.py claimed age was ruled out because an image from
2025-11 was saved while one from 2026-08 was not. That conclusion does not
follow. "Saved" means some earlier run downloaded it, not that ChatGPT
still holds it — an image captured in June looks saved forever after, even
if it died in July. A month with no losses shows the exports were timely,
not that the assets survived.

- probe_survival_by_month.py: sample images already on disk, grouped by
  their conversation's month, and ask /files/{id} whether each still
  exists today. Old months still 200 → no expiry, and cadence did not save
  them. Old months now 404 → uploads do expire and cadence is the whole
  ballgame.
- analyze_media_age.py: stop asserting the unsupported verdict; say what
  the number does and does not show, and point at the probe.

One signal there is immune to the confound and survives: 2026-07-09 kept
38 images and lost 12. Same conversation, same day, opposite outcomes —
no retention policy does that, so at least part of this is per-asset.

tools/analyze_media_age.py

@@ -20,6 +20,15 @@ grouping outcomes by month and by source distinguishes the hypotheses:
   * Neither           → deaths scattered across months, or concentrated in a
                         few conversations while their neighbours survive.
 
+IMPORTANT — what "saved" does and does not mean. It means the image was
+downloaded by *some* run at *some* point, not that ChatGPT still holds it. An
+image captured in June looks saved forever after, even if it died in July. So
+a month with no losses shows the exports were timely; it cannot show the
+assets survived. Use probe_survival_by_month.py to ask the API what is still
+live today. The one signal here that is immune to this confound is a single
+conversation that both kept and lost images: same age, same chat, opposite
+outcomes means the cause is per-asset, not retention.
+
 Run from the project root:
 
     python tools/analyze_media_age.py
@@ -123,12 +132,21 @@ def main() -> None:
         print(f"      dead range                : {old_d or '—'} … {new_d or '—'}")
         if old_s and new_d and old_s < new_d:
             print(
-                f"      → an asset from {old_s} survived while one from "
-                f"{new_d} did not: age alone does not explain the loss."
+                f"      → an asset from {old_s} was captured while one from "
+                f"{new_d} was not."
             )
+            print(
+                "        This does NOT disprove expiry: 'captured' means some "
+                "run got it in time,"
+            )
+            print(
+                "        not that it is still on ChatGPT today. Run "
+                "probe_survival_by_month.py to tell"
+            )
+            print("        those apart.")
         elif old_s and old_d and old_s > old_d:
             print(
-                f"      → everything dead is older than everything alive "
+                f"      → everything lost is older than everything captured "
                 f"(cutoff between {old_d} and {old_s}): consistent with expiry."
             )
     print()