fix(pcdb): extend heat-pump efficiency toward 100% beyond table PSR range 🟩

interpolate_heat_pump_efficiency_at_psr clamped to the smallest/largest PSR
row when the dwelling's plant size ratio fell outside the record's range.
That is the SAP 10.2 Appendix N rule for *combined heat-pump-and-boiler*
packages, not for a plain air/ground/water source heat pump.

Per Appendix N2 (PDF p.101, footnotes 44/45) a source heat pump whose PSR
exceeds the record's largest value takes a reciprocal-linear interpolation
between the largest-PSR efficiency and 100% at twice that PSR (100% beyond),
and 100% when the PSR is below the record's smallest value. Both the space-
and water-heating PSR-dependent efficiencies extend this way.

Effect: an oversized heat pump in a small dwelling is no longer credited the
full top-of-table COP. Accredited Elmhurst worksheet for cert 100110101713
(golden fixture case 56, PCDB 100061, PSR 3.107 over largest 2.0): (206)
334.4% -> 139.66% = Elmhurst exact. Corpus (RdSAP-21.0.1, n=1000) MAE
0.7397 -> 0.7258, within-0.5 0.7410 held; only two certs move (both
oversized-PSR heat pumps), 100110101713 +18.32 -> -4.97.

Exhaust-air and combined heat-pump-and-boiler packages have different
boundary rules (straight-to-100% / clamp-to-edge) but are not distinguished
by the current PCDB parse; the air/ground/water rule is applied uniformly,
a documented limitation noted in the function docstring.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

domain/sap10_calculator/tables/pcdb/parser.py

@@ -263,6 +263,14 @@ _HP_PSR_GROUP_OFFSET_PSR: Final[int] = 0
 _HP_PSR_GROUP_OFFSET_ETA_SPACE_1: Final[int] = 2
 _HP_PSR_GROUP_OFFSET_ETA_WATER_3: Final[int] = 6
 
+# SAP 10.2 Appendix N2 (PDF p.101, footnotes 44/45): out-of-range PSR
+# extension for air/ground/water source heat pumps. Above the record's
+# largest PSR the efficiency is reciprocal-interpolated toward 100% at
+# `_EXTENSION_PSR_MULTIPLE` × the largest PSR; below the smallest PSR, and
+# beyond that multiple, the efficiency is the terminal 100%.
+_EXTENSION_TERMINAL_EFFICIENCY_PCT: Final[float] = 100.0
+_EXTENSION_PSR_MULTIPLE: Final[float] = 2.0
+
 
 def _parse_psr_groups(raw: tuple[str, ...]) -> tuple[PsrEfficiencyGroup, ...]:
     """Decode the variable-length PSR-dependent block of a format-465
@@ -317,28 +325,60 @@ def interpolate_heat_pump_efficiency_at_psr(
     (not their reciprocals taken from PCDB), so the η_*_pct values must
     be strictly positive — every PCDB row in the cohort satisfies this.
 
-    Per spec PDF p.100 lines 7039-7072: clamp to the smallest PSR in
-    the record when `target_psr` is below it, and to the largest when
-    above ("if the PSR is greater than the largest PSR in the database
-    record then the heat pump space and water heating fractions for the
-    largest PSR should be used, and if the PSR is less than the
-    smallest PSR in the database record then the heat pump space and
-    water heating fractions for the smallest PSR should be used").
+    Out-of-range PSR (spec PDF p.101, footnotes 44/45 — air/ground/water
+    source heat pumps):
+
+      - Below the smallest PSR in the record: "an efficiency of 100%
+        should be used if the PSR is less than the smallest value in the
+        database record."
+      - Above the largest PSR in the record: "an efficiency may be
+        obtained from linear interpolation between that at the largest
+        PSR in the data record and efficiency 100% at PSR two times the
+        largest PSR in the data record. If the PSR is greater than two
+        times the largest PSR in the data record an efficiency of 100%
+        should be used." The interpolation is reciprocal-linear too
+        (footnote 43), with 100% as the upper anchor.
+
+    Both space- and water-heating PSR-dependent efficiencies extend the
+    same way. (Exhaust-air heat pumps and combined heat-pump-and-boiler
+    packages instead use 100% directly above the largest PSR, and combined
+    packages clamp to the edge rows; neither is distinguished by the
+    current PCDB parse, so the air/ground/water rule is applied uniformly
+    — a documented limitation. The dominant RdSAP cohort is air source.)
 
     Cohort fixture: cert 3336-2825-9400-0512-8292 (Mitsubishi PUZ-WM50VHA,
     PCDB 104568) — PSR 1.40151 brackets PCDB rows PSR 1.2 (η_space_1
     = 253.9) and PSR 1.5 (η_space_1 = 229.2). Linear (pre-slice):
     237.31; reciprocal (spec-faithful): 236.74 — matches worksheet
     (206)/(210) at 1e-4 once the 0.95 in-use factor is applied.
+
+    Out-of-range anchor: PCDB 100061 (golden fixture case 56), largest PSR
+    2.0 (η_space_1=352.0). At dwelling PSR 3.10665 the extension to 100%
+    at PSR 4.0 gives η_space_1 = 147.011 → (206) = 139.660, matching the
+    accredited Elmhurst worksheet (vs the old clamp's 352.0 → 334.4%).
     """
     if not psr_groups:
         raise ValueError("PSR groups required for interpolation")
-    if target_psr <= psr_groups[0].psr:
-        first = psr_groups[0]
-        return (first.eta_space_1_pct, first.eta_water_3_pct)
-    if target_psr >= psr_groups[-1].psr:
+    if target_psr < psr_groups[0].psr:
+        return (_EXTENSION_TERMINAL_EFFICIENCY_PCT, _EXTENSION_TERMINAL_EFFICIENCY_PCT)
+    if target_psr > psr_groups[-1].psr:
         last = psr_groups[-1]
-        return (last.eta_space_1_pct, last.eta_water_3_pct)
+        upper_psr = _EXTENSION_PSR_MULTIPLE * last.psr
+        if target_psr >= upper_psr:
+            return (
+                _EXTENSION_TERMINAL_EFFICIENCY_PCT,
+                _EXTENSION_TERMINAL_EFFICIENCY_PCT,
+            )
+        t = (target_psr - last.psr) / (upper_psr - last.psr)
+        eta_space_1 = 1.0 / (
+            (1.0 - t) / last.eta_space_1_pct
+            + t / _EXTENSION_TERMINAL_EFFICIENCY_PCT
+        )
+        eta_water_3 = 1.0 / (
+            (1.0 - t) / last.eta_water_3_pct
+            + t / _EXTENSION_TERMINAL_EFFICIENCY_PCT
+        )
+        return (eta_space_1, eta_water_3)
     for low_group, high_group in zip(psr_groups, psr_groups[1:]):
         if low_group.psr <= target_psr <= high_group.psr:
             span = high_group.psr - low_group.psr