fix(release): Acknowledge the six stale bench trailers that stopped 0.4.10-rc.1

[aywrite]

The 0.4.10-rc.1 release run (38674518239) failed at its bench pin check. Six of PR #392's protocol fixes state 9939754 while their trees pin 10026152. They were written on v0.4.9 and rebased onto the repetition fix (9e649ca) before they landed. None of them touches the pinned test, so the trailers are stale and the pins are right.

They go on the acknowledgement list, beside the six lapses of the same kind already there, and the list's test names them. check_bench_pins.py v0.4.9 HEAD now exits 0, and the script's tests pass.

The rc.1 tag stays on the failing commit. rc.2 will be cut once this merges.

🤖 Generated with Claude Code

https://claude.ai/code/session_01N1vwXYg92KXXHEn4xoRz1k

---

Generated by Claude Code

[github-actions]

Speed against 6724de82

Measured on AMD EPYC 7763 64-Core Processor.
Both sides built with rustc 1.99.0 (b940084d7 2026-09-28).

Both sides built and run on this runner in this job, the way
scripts/speed.sh measures a perf commit. Each round runs
both sides on a layout of its own, the same compiled code with
its code and data shuffled and moved, so the 95% interval carries
where the code landed as well as the run. The verdict holds the
interval against a 1% threshold. The default layout, the one a
release ships, is measured after as a diagnostic. The node
counts are the search's: they move when the search does, and a
speed change leaves them alone.

round     base nps  candidate nps  change
    1      4876550        4816787   -1.2%
    2      4726472        4771782   +1.0%
    3      4825685        4742639   -1.7%
    4      4826551        4873126   +1.0%
    5      4826928        4751932   -1.6%
    6      4808300        4871481   +1.3%
    7      4737918        4807518   +1.5%
    8      4724471        4802686   +1.7%
    9      4839259        4883778   +0.9%
   10      4885282        4851533   -0.7%
   11      4857450        4837851   -0.4%
   12      4812894        4811886   -0.0%
   13      4851176        4840006   -0.2%
   14      4772980        4846230   +1.5%
   15      4777178        4741118   -0.8%
   16      4820120        4632045   -3.9%
   17      4832201        4806632   -0.5%
   18      4786867        4813985   +0.6%
   19      4801382        4845749   +0.9%
   20      4773902        4841926   +1.4%
   21      4792532        4841768   +1.0%
   22      4768685        4843449   +1.6%
   23      4825433        4852766   +0.6%
   24      4876941        4912546   +0.7%
   25      4871694        4833159   -0.8%
   26      4715947        4852205   +2.9%
   27      4749599        4712717   -0.8%
   28      4929226        4956866   +0.6%
   29      4815923        4739557   -1.6%
   30      4797593        4695189   -2.1%
   31      4786446        4711320   -1.6%
   32      4673843        4752542   +1.7%
   33      4808690        4891297   +1.7%
   34      4786021        4818107   +0.7%
   35      4826473        4864429   +0.8%
   36      4769584        4851085   +1.7%
   37      4832127        4793594   -0.8%
   38      4661778        4803212   +3.0%
   39      4709487        4724875   +0.3%
   40      4914788        4829424   -1.7%

              nodes    time  median nps  faster half
base       11916992  2.48 s     4808495      4847770
candidate  11916992  2.47 s     4823766      4861034
change                            +0.3%        +0.3%

paired change +0.3%, 95% interval -0.2% to +0.8%
diagnostic, on the default layout alone -0.6%, 95% interval -1.5% to +0.3%, 13 rounds

no change beyond ±1.0%: the whole interval is inside it

Speed: +0.3% (bench nps, 95% interval -0.2% to +0.8%, 40 interleaved rounds over shuffled layouts vs 6724de82)

The instructions each side's bench executed, counted under
cachegrind. The count repeats to within a few hundred
instructions, so a small change here is a real one, but it
prices instructions only: cache misses, mispredicted branches
and where the code lands are the speed's to show.

instructions against 6724de82, one cachegrind run a side, bench

             instructions     nodes  per node
base       15,968,841,487  11916992    1340.0
candidate  15,968,841,516  11916992    1340.0
change             +0.00%              +0.00%

within ±0.7%, as far as edits not made for speed have moved the count