reducer path to strengthen too

docs/core_state_mapping_and_merge.md

@@ -249,14 +249,17 @@ Reducers are a capability family, similar to reusable node specs:
 - dependency-trackable
 
 State fields reference reducers declaratively. String reducer names are accepted
-as shorthand for unconfigured reducers; configured reducers use a `name` plus
-JSON-compatible `config`. Workflow artifacts do not embed arbitrary Python
-callables.
+as shorthand for unconfigured reducers; configured reducers use a structural
+`ref` plus JSON-compatible `config`. Workflow artifacts do not embed arbitrary
+Python callables.
 
 ```python
 StateField(
     type="integer",
-    reducer={"name": "wf.std.modulo_add", "config": {"modulus": 10}},
+    reducer={
+        "ref": {"source": "wf.std", "capability_key": "modulo_add"},
+        "config": {"modulus": 10},
+    },
 )
 ```
 

docs/structural_refs.md

@@ -159,16 +159,16 @@ being split again.
 Reducer refs are capability refs, not graph paths. `wf.std.add` is shorthand for
 source `wf.std` and capability key `add`.
 
-The reducer cleanup should move `ReducerRef` toward structural `CapabilityRef`
-while keeping string reducer names as parse-only shorthand. Reducer config stays
-part of the reducer reference payload:
+Configured reducer refs now use a structural `CapabilityRef` while keeping
+string reducer names as parse-only shorthand and display keys. Reducer config
+stays part of the reducer reference payload:
 
 ```json
 {
-  "name": "wf.std.modulo_add",
+  "ref": {"source": "wf.std", "capability_key": "modulo_add"},
   "config": {"modulus": 10}
 }
 ```
 
-That future cleanup must not reuse graph path parsing rules. Reducer names live
-in the capability/source domain.
+That shape must not reuse graph path parsing rules. Reducer names live in the
+capability/source domain.