fix: normalize G as scope object and add CoverageMetric(G) guardrail

FPF-Spec.md

@@ -4027,7 +4027,9 @@ Cross‑context transfer maps a **set** `Scope` via a **Bridge** and penalizes *
 What teams often want to “compress into a G number” is actually (a) the **quality of expression** and (b) the **completeness** of the declared scope. The first is an **F threshold** (e.g., require **`U.Formality ≥ F4`** so the scope is predicate‑like and addressable); the second is handled by explicit **ESG guards**: “**Scope covers TargetSlice**,” “**`Γ_time` is specified**,” and “**freshness window holds**” (R‑lane). A ladder for G adds confusion but no additional control.
 
 **Normative directive.**
-`U.ClaimScope (G)` **SHALL** remain a **set‑valued** characteristic; **no ordinal or numeric ladder SHALL be defined** for G. Authoring and gating **SHOULD** use **F thresholds** (C.2.3) and **explicit guard predicates** (A.2.6) rather than pseudo‑levels of G.
+`U.ClaimScope (G)` **SHALL** remain a **set‑valued scope object**; **no ordinal or numeric ladder SHALL be defined** for G. Authoring and gating **SHOULD** use **F thresholds** (C.2.3) and **explicit guard predicates** (A.2.6) rather than pseudo‑levels of G.
+
+**Report-only proxy (optional).** `CoverageMetric(G)` MAY be emitted for dashboards/reports as a derived scalar projection of `G` (with declared domain/measure references). It **MUST** be marked report-only and **MUST NOT** substitute `G` in norms, gate predicates, or conformance logic.
 
 #### A.2.6:7.5 - Translation across Contexts (Bridge & CL)
 
@@ -23557,7 +23559,7 @@ Every non‑trivial result in FPF—*a composed system is safe*, *a model is cre
 
 To make such claims comparable and auditable across domains, B.3 introduces a **Trust & Assurance Calculus** that:
 
-* uses a **small set of characteristics** (F–G–R) governed by CHR principles (these are **not** a state space),
+* uses **two characteristics (F,R) + one scope object (G)**; only **F/R** are CHR characteristics (this is **not** a state space),
 * accounts for **integration quality** via **Congruence Level (CL)** along the edges of a `DependencyGraph` (B.1.1, A.14),
 * and composes these values with **Γ‑flavours** while respecting the **Invariant Quintet** (IDEM, COMM/LOC or their replacements, WLNK, MONO).
 
@@ -23592,9 +23594,9 @@ Without a disciplined calculus, four chronic failures appear:
 
 B.3 defines **what** to measure, **how** those measures live on nodes and edges of the dependency graph, and the **shape** of the aggregation that any Γ‑flavour must honor when producing an *assurance result*.
 
-#### B.3:4.1 - The F–G–R characteristics (CHR‑compliant)
+#### B.3:4.1 - The F–G–R assurance components (two CHR characteristics + one scope object)
 
-We standardize three characteristics on **nodes (holons)** plus one **edge** characteristic:
+We standardize two **node characteristics** (**F**, **R**), one **node scope object** (**G**), plus one **edge** characteristic (**CL**):
 
 1. **Formality (F)** — *how constrained the reasoning is by explicit, proof‑grade structure.*
 
@@ -23605,15 +23607,17 @@ We standardize three characteristics on **nodes (holons)** plus one **edge** cha
 
 2. **ClaimScope (G)** — *how broadly the result applies in the relevant domain space.*
 
-   * **Scale kind:** **coverage / span** (set‑ or measure‑based; domain‑specific).
-   * **Monotone direction:** larger, but only when **correctly supported** (see WLNK and CL below).
+   * **Kind:** **set‑valued scope object** over `U.ContextSlice` (USM set algebra; not a CHR scale).
+   * **Order semantics:** set inclusion / coverage predicates only; “larger is better” is **not** a universal invariant under bridge translation.
+   * **Optional report-only projection:** `CoverageMetric(G)` MAY be published as a scalar proxy for reporting, but it MUST be explicitly derived from `G` with declared domain/measure references.
+   * **Normative guardrail:** `CoverageMetric(G)` **MUST NOT** replace `G` in normative clauses, gate conditions, or conformance checks.
 
 3. **Reliability (R)** — *how likely the claim/behavior holds under stated conditions.*
 
    * **Scale kind:** **ratio** in `[0,1]` (or a conservative ordinal proxy when numeric modeling is unavailable).
    * **Monotone direction:** higher is better.
 
-2. **Congruence Level (CL)** — *edge property: how well two parts fit* (semantic alignment, calibration, interface Standard).
+4. **Congruence Level (CL)** — *edge property: how well two parts fit* (semantic alignment, calibration, interface Standard).
 
    * **Scale kind:** **ordinal** with a **monotone penalty function** `Φ(CL)` where `Φ` decreases as CL increases.
    * **Canonical levels (example):**
@@ -32615,7 +32619,7 @@ The FPF ecosystem is formally stratified into three canonical **artefact familie
     *   **Pedagogical Companion:** Provides a tutorial on how to model a water pump using that profile.
 
 *   **For an `U.Episteme`:**
-    *   **Conceptual Core:** Defines `U.Episteme` and the F-G-R characteristics.
+    *   **Conceptual Core:** Defines `U.Episteme` and the F-G-R components.
     *   **Tooling Reference:** Provides the reference linting tool to automatically score epistemes.
     *   **Pedagogical Companion:** Provides a case study on how a scientific theory's R-score evolves over time.