docs: say what the graphed hooks do without the project's labels

The hist.graphed docstring and the test-module header carried graphed-project
work-item and freeze-tag labels an upstream reader cannot resolve; they now state
the same behaviour in plain words.

Assisted-by: ClaudeCode:claude-opus-5

Author: lgray

src/hist/graphed/hist.py

@@ -52,6 +52,6 @@ def fill(
 class Hist(FillModeMixin, HistInMemory[S], ghb.Histogram, Generic[S], family=hist):  # type: ignore[misc]
     """A `hist.Hist` whose fills are DEFERRED graphed computations: QuickConstruct
     (`Hist.new.Reg(...).Double()`) and named-axis fills record into the graphed IR. Evaluation
-    is graphed's own idiom — `plan()` + an R7 executor (whose result wraps back into an
+    is graphed's own idiom — `plan()` + a graphed executor (whose result wraps back into an
     in-memory `hist.Hist` via `hist.Hist(value)`), or the reference `session.materialize` on a
     fill node."""

tests/test_graphed.py

@@ -1,9 +1,9 @@
-# Tests for hist.graphed — deferred hist filling on graphed task graphs (HIST-1; P0.1 of the
-# ADL-benchmarks port). QuickConstruct-built deferred histograms must equal their eager hist.Hist
-# twins BIT FOR BIT over graphed-numpy AND graphed-awkward sources and over a real uproot TTree,
-# with named-axis fills, weights, NamedHist, and the partition-wise efficiency witness (the
-# source's whole-dataset loader never runs). Evaluation is graphed's idiom [freeze-HIST-2,
-# user-directed]: plan() + an R7 executor; hist.Hist(value) wraps results back in-memory.
+# Tests for hist.graphed — deferred hist filling on graphed task graphs. QuickConstruct-built
+# deferred histograms must equal their eager hist.Hist twins BIT FOR BIT over graphed-numpy AND
+# graphed.awkward sources and over a real uproot TTree, with named-axis fills, weights, NamedHist,
+# and the partition-wise efficiency witness (the source's whole-dataset loader never runs).
+# Evaluation is graphed's idiom: plan() + a graphed executor; hist.Hist(value) wraps results back
+# in-memory.
 from __future__ import annotations
 
 import numpy as np