docs(telemetry): remove pre-squash references from the gap-fill comments

These comments dated themselves against this branch or against an earlier
revision of the same change, neither of which survives a squash merge.

- 'as of this branch' in the pricing-case doc and the pass-through static_assert
  becomes a statement about what the constants currently produce.
- 'Extracted from processGetObjectByHash()' and 'Split from start()' describe
  edits internal to this change; both now say why the method stands alone.
- Recording.h: the mock-abstract mismatch is a standing consequence of a member
  set that differs between builds, not something that 'has previously' happened.
- MetricsRegistry.cpp: describe the loops.txt entry as recording two cycles
  rather than as what ordering.txt 'previously had'.
- InboundLedger.h: the acquire span is the only signal for back-fill cost; it
  did not 'previously emit' nothing.
- check_bucket_parity.py: replace the eleven-phase drift story with the reason
  the check exists, and point the failure message at HistogramBuckets.h and the
  collector config instead of OpenTelemetryPlan/, which does not reach develop.

Comments and one error message only, no behaviour change.

.github/scripts/telemetry/check_bucket_parity.py

@@ -2,13 +2,13 @@
 """Assert the C++ millisecond ladder agrees with the collector's spanmetrics ladder.
 
 The two are specified to match so a span-derived latency panel and a native
-histogram panel can be read on the same scale. They *were* identical when first
-shipped. Then the collector ladder alone was extended -- sub-millisecond edges
-below 1ms and second-scale edges up to 30s -- and nothing checked the other
-side, so the C++ ladder stayed capped at 5s. Every quantile above 5s then read
-back as a flat 5000, because Prometheus returns the second-highest edge for a
-quantile landing in the `+Inf` bucket. That looks like a measurement rather
-than an error, which is why it survived for eleven phases.
+histogram panel can be read on the same scale. Nothing else couples them, so
+extending one ladder alone -- sub-millisecond edges below 1ms, second-scale
+edges up to 30s -- silently leaves the other short. That failure is quiet:
+Prometheus returns the second-highest edge for a quantile landing in the
+`+Inf` bucket, so every quantile above a too-low ceiling reads back as a flat
+number that looks like a measurement rather than an error. This check is what
+makes the drift loud.
 
 The rule is containment, not equality:
 
@@ -17,7 +17,7 @@ The rule is containment, not equality:
   * the C++ ladder MAY carry extra edges ABOVE the collector's highest edge,
     because jobs outlive spans -- the updatepaths job type was measured
     averaging ~60s, which no span approaches. Demanding equality would force a
-    ceiling that censors it, reintroducing the bug this guards against;
+    ceiling that censors it, recreating the failure this guards against;
   * collector edges below 1ms are expected to be ABSENT rather than missing:
     beast::insight::Event rounds every duration up to a whole millisecond
     before it reaches the histogram, so those edges could never collect a
@@ -117,8 +117,10 @@ def main():
         )
     print(
         "\nThe two ladders must agree over their shared range. Extra C++ edges are\n"
-        "permitted only ABOVE the collector's highest edge. Change both sides, or\n"
-        "change the spec in OpenTelemetryPlan/Phase7_taskList.md.",
+        "permitted only ABOVE the collector's highest edge. To re-price the shared\n"
+        f"range, edit the ladder in {HEADER} and the\n"
+        f"spanmetrics 'buckets:' list in {COLLECTOR}\n"
+        "in the same change, so both sides stay in step.",
         file=sys.stderr,
     )
     return 1

include/xrpl/telemetry/Recording.h

@@ -6,12 +6,12 @@
  *  Each type below holds real state when telemetry is compiled in and is an
  *  empty type with no-op methods when it is not. The member is declared in
  *  both configurations, so a class's member set and public API never differ
- *  between builds -- a difference that has previously made a test mock
- *  abstract. The compiled-out forms are empty types, so such a member costs a
- *  byte of padding rather than nothing. `[[no_unique_address]]` would remove
- *  even that, but MSVC ignores the standard spelling for ABI compatibility, so
- *  it is deliberately not used. What these types buy is work not being done,
- *  not a smaller struct.
+ *  between builds -- a difference that leaves a test mock complete in one
+ *  configuration and abstract in the other, where it then fails to compile.
+ *  The compiled-out forms are empty types, so such a member costs a byte of
+ *  padding rather than nothing. `[[no_unique_address]]` would remove even that, but MSVC ignores
+ *  the standard spelling for ABI compatibility, so it is deliberately not
+ *  used. What these types buy is work not being done, not a smaller struct.
  *
  *      kEnabled ---- if constexpr ---- telemetry-only blocks
  *          |

src/test/overlay/TMGetObjectByHash_test.cpp

@@ -723,9 +723,9 @@ class TMGetObjectByHash_test : public beast::unit_test::Suite
      * One pricing case: inputs, the derived expectation, and the literal.
      *
      * Both expectations are kept. `derived` is written from the Tuning
-     * constants so a deliberate re-pricing needs one edit; `literal` is the
-     * number as of this branch so a re-pricing cannot pass unnoticed by
-     * being self-consistently wrong.
+     * constants so a deliberate re-pricing needs one edit; `literal` pins the
+     * number those constants currently produce, so a re-pricing cannot pass
+     * unnoticed by being self-consistently wrong.
      */
     struct FeeCase
     {

src/tests/libxrpl/telemetry/MetricsRegistry.cpp

@@ -204,9 +204,9 @@ allFoldToOther()
 static_assert(allPassThroughUnchanged());
 static_assert(allFoldToOther());
 
-// The verified size of the pass-through set as of this branch: 43 all-letter
-// job-name literals. Pinned so that adding or removing a job name without
-// revisiting the label-cardinality budget fails the build here.
+// The pass-through set is every all-letter job-name literal in the tree: 43
+// of them. Pinned so that adding or removing a job name without revisiting
+// the label-cardinality budget fails the build here.
 static_assert(kPassThroughHandlers.size() == 43);
 
 /**

src/xrpld/app/ledger/InboundLedger.h

@@ -226,8 +226,8 @@ private:
     /**
      * Spans the acquire lifecycle: started in init(), finalized in done()
      * with the outcome (complete/failed), timeout count, and peer count.
-     * Gives operators visibility into back-fill / fork-recovery cost, which
-     * previously emitted no span or metric.
+     * This span is the only signal for back-fill / fork-recovery cost; no
+     * other span or metric covers it.
      * Thread-free: emplaced by the acquiring thread, reset on a JtLedgerData
      * worker. A SpanGuard owns no thread-local Scope, so it can be destroyed
      * on the worker without corrupting the origin thread's context stack.

src/xrpld/overlay/detail/PeerImp.h

@@ -704,10 +704,10 @@ private:
     /**
      * Record the OTel metrics for one completed `TMGetObjectByHash` request.
      *
-     * Extracted from `processGetObjectByHash()` purely to keep that method
-     * within the 80-line limit; it holds no logic of its own beyond deriving
-     * the hit/miss split from `requested` and `found`. Called once per
-     * request, after the fetch loop and the `charge()` call.
+     * Called once per request from `processGetObjectByHash()`, after the fetch
+     * loop and the `charge()` call. A separate method so that one stays within
+     * the 80-line limit; it holds no logic of its own beyond deriving the
+     * hit/miss split from `requested` and `found`.
      *
      * Records `getobject_request_objects`, `getobject_lookup_us`,
      * `getobject_charge`, and both label values of

src/xrpld/telemetry/MetricsRegistry.cpp

@@ -30,11 +30,12 @@
 
 // The app and overlay includes below are why
 // .github/scripts/levelization/results/loops.txt records
-// `xrpld.app <-> xrpld.telemetry` and `xrpld.overlay <-> xrpld.telemetry`, where
-// ordering.txt previously had telemetry strictly below both. The observable
-// gauges are pull-model: their callbacks sample live state when the reader
-// thread fires, so they need the concrete types to call getJqTransOverflow(),
-// size(), getPeerDisconnectCharges(), foreach() and txMetrics().
+// `xrpld.app <-> xrpld.telemetry` and `xrpld.overlay <-> xrpld.telemetry` as
+// cycles, rather than an acyclic ordering.txt entry placing telemetry strictly
+// below both. The observable gauges are pull-model: their callbacks sample live
+// state when the reader thread fires, so they need the concrete types to call
+// getJqTransOverflow(), size(), getPeerDisconnectCharges(), foreach() and
+// txMetrics().
 //
 // The cycle is confined to this translation unit. No telemetry header includes
 // app or overlay (MetricsRegistry.h forward-declares what it needs and takes a

src/xrpld/telemetry/MetricsRegistry.h

@@ -308,11 +308,11 @@ public:
      * phase. Mostly ObservableGauges, plus the ObservableCounters whose
      * source value is already cumulative.
      *
-     * Split from `start()` because the two halves have different
-     * prerequisites. `start()` needs only config strings; these callbacks
-     * read live Application services, so this half must run later.
+     * A separate entry point from `start()` because the two halves have
+     * different prerequisites. `start()` needs only config strings; these
+     * callbacks read live Application services, so this half must run later.
      * Registering an observable also arms the reader thread to invoke its
-     * callback on the next tick, which is why the split is about ordering
+     * callback on the next tick, which is why the separation is about ordering
      * and not just tidiness.
      *
      * @pre `start()` has already run (the meter exists). If it has not,