test: prune duplicate tests, re-sight two blind coverage guards

Follow-up to a fan-out audit of the whole test suite (742 tests, 21
read-only assessors + synthesis). Two clusters of the audit were
actioned here; the #66 hashing-removal cohort it surfaced is left for
the #66 scope, and the RED-first doc-rot sweep is deferred.

§2 — true duplicates removed (each kept test is a strict superset or
identical fixture+assertions of the deleted one):
- ailang-check param_in_reject_message_names_allowed_set (kept the
  _renders_allowed_set_in_ailang_syntax superset)
- ailang-check cross_module_pat_ctor_typedriven_resolves (kept ct2_;
  shared cross_module_ws helper retained, 2 other callers)
- ailang-core mq1_qualified_instancedef_class_accepted (kept ct1_)
- ail e2e bool_to_str_emits_true_branch (kept the rc-stats superset)
- ail loop_recur_str adt-leg guard (the standalone heap pin owns it;
  dropped the dangling header reference too)
- ailang-check over_strict_mode_silent_when_param_is_borrow (its
  assertion is subsumed by explicit_fn_with_no_uses_is_clean; its own
  doc claimed a "borrowing body" the Lit body never provided)

§4 — meta-guards that had gone partially blind re-sighted:
- spec_drift::spec_mentions_every_def_kind now pushes Class/Instance
  exemplars and asserts their `(class `/`(instance` anchors; the
  "wait for 22b.4" exclusion was stale (anchors shipped in e809f45).
- schema_coverage now tracks Term::New (VariantTag + EXPECTED +
  visitor insert); the new_*/raw_buf_* fixtures already exercise it,
  so the "no fixture emits new" exclusion was stale.
- prose doc_wrap_widow_control_skips_when_combined_too_long was
  tautological: its 32/32/42-char words gave combined=75<=80, so
  widow control actually fired. Rebuilt the fixture (short head + two
  40-char words → combined=81>80) so it hits the real skip branch,
  and added the missing "last line stays 1 word" assertion. The
  obvious 2-word fixture does NOT work — it skips via the
  prev_words<2 branch, not the combined-over-budget branch.

ailang-check 124->121, ailang-core 55->54, e2e 102->101,
loop_recur_str 4->3; spec_drift 8 green, schema_coverage green,
prose lib 63 green.

crates/ailang-core/src/workspace.rs

@@ -2638,40 +2638,6 @@ mod tests {
         }
     }
 
-    /// a qualified `InstanceDef.class` referencing a class in
-    /// an imported module is the canonical form post-canonical-class-form and is
-    /// accepted by `validate_canonical_type_names`. Symmetric to the canonical-form rule's
-    /// "qualified Type::Con resolves" positive path.
-    #[test]
-    fn mq1_qualified_instancedef_class_accepted() {
-        let a: Module = serde_json::from_value(serde_json::json!({
-            "schema": crate::SCHEMA,
-            "name": "A",
-            "imports": [],
-            "defs": [{
-                "kind": "class",
-                "name": "Show",
-                "param": "a",
-                "methods": []
-            }]
-        })).unwrap();
-        let b: Module = serde_json::from_value(serde_json::json!({
-            "schema": crate::SCHEMA,
-            "name": "B",
-            "imports": [{ "module": "A" }],
-            "defs": [{
-                "kind": "instance",
-                "class": "A.Show",
-                "type": { "k": "con", "name": "Int" },
-                "methods": []
-            }]
-        })).unwrap();
-        let mut modules = BTreeMap::new();
-        modules.insert("A".to_string(), a);
-        modules.insert("B".to_string(), b);
-        validate_canonical_type_names(&modules, &["prelude"]).expect("qualified A.Show must be accepted");
-    }
-
     /// pd.1 Task 1: `load_modules_with` returns modules without injecting
     /// prelude. The prelude inject is the caller's responsibility (the
     /// shim `load_workspace_with` does it in pd.1; surface owns it from