Refresh design-backlog docs after Pot's gain/theta cleanup

Drop the now-stale "config.json.templ/.sim still carry dead
Model.kn/Plant.kn keys" note (they were deleted), note that Pot now
always starts at theta_amb (no separate initial-temp param) and that
dropping the gain (efficiency) factor entirely is a partial realism
regression, not just a fix. Point the config-validation item at the
gain key that's still dead in config.json.sim. Also update demo_sud.py
to match the Pot params Pot.__init__ actually reads.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

components/plant/TODO.md

@@ -22,29 +22,37 @@ the current `delay`-line rewrite of `Pot`.
   wired up as plant-side measurement noise. Sensor noise is now
   modeled one layer up, in `TempSensorSim` (`temp_offset`/`variance`,
   wired from `brewpi.py`'s `TempSensorFactory.create(..., temp_offset=
-  -0.15, variance=0.01)`), not in the plant model. Note: `config.json.templ`
-  and `config.json.sim` still carry stale `Model.kn`/`Plant.kn` keys
-  that nothing reads anymore — harmless but worth deleting (see
-  `components/pid/TODO.md`'s config-validation item).
+  -0.15, variance=0.01)`), not in the plant model. The stale
+  `Model.kn`/`Plant.kn` keys this item flagged have since been deleted
+  from `config.json.templ`/`config.json.sim` too.
 
 - [x] **`theta_amb` is hardcoded, not configured.** Fixed: added a
   top-level `ambient_temperature` key to `config.json`(`.templ`/`.sim`);
   `brewpi.py` reads it once and passes it to both the real `Pot` and
   (via a new `theta_amb` param) `temp_controller_smith.py`'s internal
   models, so the Smith predictor's model matches the real plant's
-  assumed ambient instead of silently defaulting to 20. Real ambient
-  temperature still isn't modeled as a *drifting* disturbance (it's a
-  static config value, not a live sensor reading), so robustness to
-  ambient drift remains untested — that part of the original concern
-  stands.
+  assumed ambient instead of silently defaulting to 20. `Pot` also
+  dropped its separate `params['theta']` initial-temperature field —
+  it now always starts at `theta_amb` (so the same config value governs
+  both the assumed ambient *and* the pot's starting temperature; there's
+  no longer a way to start a sim at a temperature other than ambient).
+  Real ambient temperature still isn't modeled as a *drifting*
+  disturbance (it's a static config value, not a live sensor reading,
+  though `Pot.set_ambient_temperature()` now exists as a hook for that),
+  so robustness to ambient drift remains untested.
 
-- [ ] **No actuator/process realism.** Partially improved: `process()`
-  now clamps `self.temp` to `min(100, ...)` (won't simulate past
-  boiling), but there's still no heater power saturation enforced
-  inside `Pot` itself (handled ad hoc in demo harnesses), no
-  evaporation/lid heat loss, and no varying thermal mass `M` as volume
-  changes (e.g. boil-off, adding water/grain) — `M` is a fixed
-  constant for the whole simulation.
+- [ ] **No actuator/process realism.** Partially improved (`process()`
+  clamps `self.temp` to `min(100, ...)`, won't simulate past boiling)
+  and partially regressed: `Pot` also dropped its `gain` (heater
+  efficiency) factor entirely — `process()` now applies 100% of
+  commanded power with no loss, whereas before `params['gain']` could
+  model e.g. an 80% efficient transfer. `Model.gain`/`Plant.gain` are
+  now dead keys in `config.json.sim` (already absent from `.templ`) —
+  see `components/pid/TODO.md`'s config-validation item. Still no
+  heater power saturation enforced inside `Pot` itself (handled ad hoc
+  in demo harnesses), no evaporation/lid heat loss, and no varying
+  thermal mass `M` as volume changes (e.g. boil-off, adding
+  water/grain) — `M` is a fixed constant for the whole simulation.
 
 - [x] **Debug `print()`s left in `Pot.process()`.** Fixed: the
   unconditional `p_loss`/`theta_amb`/`temp` prints (leftover from the