fix(sat): fan check uses hottest GPU load + IPMI-hang-proof polling

- GPU load switches from bee-gpu-burn (compute burn, ~88% TDP) to
  dcgmproftester -t 1004 / targeted_power via
  resolveBenchmarkPowerLoadCommand — the same engine Power/Thermal Fit
  uses and the hottest sustained NVIDIA load we have, so fans are
  actually pushed toward their ceiling.
- Sample loop is now IPMI-hang-proof: every ipmitool read is time-boxed
  in an abandonable goroutine, and the poll interval backs off
  geometrically (1s→30s) when reads are slow, tightening again on
  recovery. A plateau is only trusted while telemetry is healthy;
  degraded runs ride out to MaxLoadSec. Summary gains fan_samples /
  telemetry_degraded. Drops the per-second nvidia-smi+power+cpu-temp
  sampling from the hot loop.
- Dead code removed: FanStressRow, GPUStressMetric, sampleFanStressRow,
  sampleGPUStressMetrics, WriteFanStressCSV/WriteFanSensorsCSV,
  analyzeMaxTemp, sampleSystemPowerResolved.

Topology fan tiles:
- size encodes the fan's ceiling RPM (its class), not current speed;
  coloured fill rising from the bottom encodes live duty cycle
  (current / ceiling), shown only when the ceiling was measured.
- glyph spin rate now maps absolute RPM into a human-perceptible band
  (fanSpinPeriodSec: 2.2s/turn at <=1000 RPM, 0.35s at >=13000).
- the "N fans · N OK · tile size ∝ …" caption line is gone.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019VHG21rgTUiR1G3qFHTVmN

bible-local/README.md

@@ -76,7 +76,7 @@ Generic engineering rules live in `bible/rules/patterns/`.
 - GPU bandwidth check
   - `dcgmi diag -r nvbandwidth` (per CPU socket, then all selected GPUs, on multi-socket systems -- see `decisions/2026-07-27-nvbandwidth-per-socket-split.md`)
 - Fan ceiling check (Load tier / `3. Load` only)
-  - `stressapptest` (or `stress-ng`) + `bee-gpu-burn` / `rvs gst` at 100%, run simultaneously
-  - `ipmitool sdr type Fan` sampled 1 Hz until every fan plateaus (~1 min flat) or the 15 min cap
-  - records each fan's observed peak RPM to `/var/log/bee-sat/fan-observation.json` (used by the Topology fan tiles); FAIL only on a fan at 0 RPM / IPMI cr-nr under load; **cancelled ("not applicable")**, never failed, if the host cannot be loaded or has no fan sensors
+  - `stressapptest` (or `stress-ng`) + the hottest GPU load — `dcgmproftester -t 1004` / `targeted_power` (Power/Thermal Fit engine), **not** `bee-gpu-burn` — run simultaneously
+  - `ipmitool sdr type Fan` on an adaptive interval (1 s floor, backs off to 30 s when the BMC gets slow, every read time-boxed) until every fan plateaus (~1 min flat, only trusted while telemetry is healthy) or the 15 min cap
+  - records each fan's observed peak RPM to the fan-observation store (used by the Topology fan tiles: size ∝ ceiling, fill ∝ duty cycle); FAIL only on a fan at 0 RPM / IPMI cr-nr under load; **cancelled ("not applicable")**, never failed, if the host cannot be loaded or has no fan sensors
   - see `decisions/2026-09-04-fan-ceiling-check.md`

bible-local/decisions/2026-09-04-fan-ceiling-check.md

@@ -29,14 +29,30 @@ resulting RPM — does not work on our current platform:
 The `fan` SAT test (`platform.RunFanCheck`, formerly the unwired
 `RunFanStressTest`) drives load, not the BMC:
 
-1. Start `stressapptest` (CPU + memory) and, when a GPU is present, a GPU burn
-   (`bee-gpu-burn` / `rvs gst`) **simultaneously**, each in its own goroutine.
+1. Start `stressapptest` (CPU + memory) and, when a GPU is present, the
+   **hottest sustained GPU load** — `dcgmproftester -t 1004` /
+   `targeted_power`, via `resolveBenchmarkPowerLoadCommand`, the same engine
+   Power/Thermal Fit uses — **simultaneously**, each in its own goroutine.
+   NOT `bee-gpu-burn`: that is a compute-throughput burn that tops out around
+   88% of TDP (measured ~525 W of 600 W on RTX PRO 6000 Blackwell) and never
+   makes the fans demand their true ceiling. A first run on the MSI stand with
+   `bee-gpu-burn` peaked F2U fans at 24 400 RPM vs a historical 26 000.
    A missing GPU is not an error — CPU/memory load alone exercises the cooling
    loop. Load sources reach full power at different times, so the plateau clock
    only starts once every launched source reports its process running (plus a
    fixed GPU ramp grace).
-2. Sample every fan once a second. Per fan, track the peak RPM and the last time
-   it climbed by more than `PlateauDeltaRPM` (default 50).
+2. Sample every fan on an adaptive interval (floor 1 s). Each read is
+   time-boxed (`readFansBounded` — a goroutine abandoned on timeout, so a
+   KCS-wedged `ipmitool` can never block the loop); when reads are slow the
+   interval backs off geometrically to 30 s and tightens again when they
+   recover. Under 8-GPU + CPU load on the MSI stand, `ipmitool sdr type Fan`
+   took ~14 s/call and at one point wedged for minutes — without this the
+   "1 Hz" sampler silently degraded to one sample per 14 s and the plateau
+   timer ran on stale data. A plateau is only declared while telemetry is
+   healthy (interval near the floor, ≥5 recent samples); a degraded run just
+   rides out to `MaxLoadSec` and records the peak it saw. Per fan, track the
+   peak RPM and the last time it climbed by more than `PlateauDeltaRPM`
+   (default 50).
 3. When no fan has climbed for `PlateauHoldSec` (default 60 s) and at least one
    fan rose meaningfully above baseline, declare the ceiling found and stop —
    **success**. `MaxLoadSec` (default 900 s) is a hard cap; hitting it is also