docs(diagnostics): R29 PAD mood-dynamics probe — 3-baseline series + step-response + findings

Live re-run of the PAD-dynamics probe for brokkr-smithy R29 (the original
per-turn series was never persisted). Captures, durably:

- 3-baseline arc (lofn/mimir/forseti) — decay anchor is neutral (0,0,0), not
  the trait set-point (all three converge to 0 proportional to distance).
- Positive/negative step-response (lofn) — geometric decay ~0.90 retention/turn,
  ~2-turn appraisal latency, ceiling-driven hedonic asymmetry.
- Findings: over-regulation is at the emotion->PAD gain, not appraisal; OCC
  emissions observable via emotions_active on base agents.

Methodology note: mask-hosted transient characters have a static mood engine;
base persona agents run appraisal->PAD. relations[] is provider-only per
Worldtree ADR-0009.

docs/diagnostics/r29-pad-step-pos-lofn.csv

@@ -0,0 +1,13 @@
+drive,sent,reflects_after,P,A,D,emotions
+1,impulse_pos,baseline,0.809,-0.153,0.248,
+2,neutral_1,impulse_pos,0.809,-0.153,0.248,
+3,neutral_2,neutral_1,0.809,-0.153,0.248,
+4,neutral_3,neutral_2,0.7667,-0.1259,0.2482,love=0.877
+5,neutral_4,neutral_3,0.7283,-0.1196,0.2358,love=0.855
+6,neutral_5,neutral_4,0.6449,-0.1068,0.2248,love=0.829; disgust=0.877
+7,neutral_6,neutral_5,0.582,-0.0964,0.2029,love=0.803; disgust=0.849
+8,neutral_7,neutral_6,0.5253,-0.087,0.1831,love=0.786; disgust=0.831
+9,neutral_8,neutral_7,0.499,-0.0826,0.174,love=0.769; disgust=0.814
+10,neutral_9,neutral_8,0.4503,-0.0746,0.157,love=0.749; disgust=0.792
+11,neutral_10,neutral_9,0.4064,-0.0673,0.1417,love=0.729; disgust=0.771
+12,settle_probe,neutral_10,0.3861,-0.064,0.1346,love=0.715; disgust=0.756