refactor: extract ModeCounter files (no contract change)

[Code0987]

Design

docs/design/2026-09-02-refactor-counter.md

Independent of #40 (HLL) and #41 (Bitmap). Same layout, Counter only.

What landed

Same Incr / CounterGet contract (peer CounterIncr still returns n).

File	Role
pkg/store/counter.go	store methods (out of memory.go)
pkg/engine/counter.go	public verbs
pkg/engine/counter_apply.go	owner write + snapshot install
pkg/engine/counter_cluster.go	GetOrLoad / snapshot fan-out

Tests

go test ./pkg/counter ./pkg/store ./pkg/engine

Local bench

BenchmarkStoreGetHit 2 allocs/op, BenchmarkEngineGetHit 15 allocs/op (flat).

Bench risk

No new Get/Peek logic. Shared smoke/micro ±10%. Get-hit allocs must stay flat.

[github-actions]

SuperCache bench vs main

Both sides ran once on this same GitHub runner (main, then this PR). Not a merge gate.

Baseline: cc4a44e (2026-09-02T13:48:42Z) → this run: cc4a44e (2026-09-02T13:47:26Z)

scbench smoke

cell	ops/s (main → PR)	Δ ops/s	p50	p99
get/hit n=1 c=1 uniform	8883 → 8834	-0.5%	112µs → 113µs	189µs → 187µs
get/hit n=1 c=10 uniform	35650 → 35009	-1.8%	256µs → 261µs	751µs → 772µs
get/hit n=3 c=10 uniform	40409 → 40304	-0.3%	219µs → 220µs	820µs → 822µs
miss/miss-cacheonly n=1 c=1 uniform	8660 → 8647	-0.2%	114µs → 115µs	209µs → 207µs
miss/miss-cacheonly n=1 c=10 uniform	36711 → 35061	-4.5%	247µs → 257µs	778µs → 816µs
set/put n=1 c=1 uniform	8527 → 8555	+0.3%	116µs → 115µs	202µs → 197µs
set/put n=1 c=10 uniform	33824 → 33900	+0.2%	271µs → 270µs	801µs → 801µs

Δ ops/s: green-ish is higher throughput. Latency: lower is better.

go test -bench (ns/op)

benchmark	ns/op (main → PR)	Δ	allocs/op
BenchmarkEngineBitCount	170 → 169	-0.1%	0 → 0
BenchmarkEngineBitGetHit	164 → 164	+0.1%	0 → 0
BenchmarkEngineBitGetHitParallel	210 → 192	-8.7%	0 → 0
BenchmarkEngineBitSet	505 → 507	+0.3%	2 → 2
BenchmarkEngineBitSetParallel	582 → 598	+2.8%	2 → 2
BenchmarkEngineCMSIncr	8444 → 8593	+1.8%	1 → 1
BenchmarkEngineCMSIncrN	8322 → 9377	+12.7%	1 → 1
BenchmarkEngineCMSIncrParallel	12421 → 15123	+21.8% worse	1 → 1
BenchmarkEngineCMSQueryHit	203 → 202	-0.3%	1 → 1
BenchmarkEngineCMSQueryHitParallel	250 → 248	-0.8%	1 → 1
BenchmarkEngineCounterGet	160 → 159	-0.3%	0 → 0
BenchmarkEngineCounterGetParallel	192 → 192	+0.3%	0 → 0
BenchmarkEngineDelete	822 → 810	-1.4%	6 → 6
BenchmarkEngineGeoAdd	9879 → 9912	+0.3%	8 → 8
BenchmarkEngineGeoCard	161 → 160	-0.5%	0 → 0
BenchmarkEngineGeoPosHit	101 → 101	-0.1%	0 → 0
BenchmarkEngineGeoRadius	344460 → 350679	+1.8%	1016 → 1016
BenchmarkEngineGetHit	1192 → 1199	+0.6%	15 → 15
BenchmarkEngineGetHitParallel	1242 → 1213	-2.3%	15 → 15
BenchmarkEngineGetMissCacheOnly	875 → 890	+1.7%	14 → 14
BenchmarkEngineGetMissCacheOnlyParallel	643 → 667	+3.7%	14 → 14
BenchmarkEngineGetMissLoadThrough	1646 → 1635	-0.7%	24 → 24
BenchmarkEngineGetMissLoadThroughParallel	1833 → 1820	-0.7%	23 → 23
BenchmarkEngineHGetAll	14719 → 15389	+4.6%	134 → 134
BenchmarkEngineHGetHit	109 → 112	+2.9%	1 → 1
BenchmarkEngineHGetHitParallel	138 → 128	-7.3%	1 → 1
BenchmarkEngineHGetMiss	173 → 173	+0.3%	0 → 0
BenchmarkEngineHLLAdd	2183 → 2630	+20.5% worse	1 → 1
BenchmarkEngineHLLAddParallel	2866 → 3429	+19.6%	1 → 1
BenchmarkEngineHLLCountHit	115812 → 115856	+0.0%	0 → 0
BenchmarkEngineHLLCountHitParallel	132820 → 133579	+0.6%	0 → 0
BenchmarkEngineHLen	160 → 160	+0.1%	0 → 0
BenchmarkEngineHSet	10879 → 10984	+1.0%	6 → 6
BenchmarkEngineHSetParallel	13426 → 11669	-13.1%	6 → 6
BenchmarkEngineIncr	380 → 383	+0.7%	1 → 1
BenchmarkEngineIncrParallel	421 → 428	+1.8%	1 → 1
BenchmarkEngineJsonGet	195 → 195	+0.0%	1 → 1
BenchmarkEngineJsonGetHit	413 → 431	+4.3%	5 → 5
BenchmarkEngineJsonGetHitParallel	486 → 500	+2.9%	5 → 5
BenchmarkEngineJsonSet	28422 → 29177	+2.7%	215 → 215
BenchmarkEngineJsonSetParallel	17669 → 17340	-1.9%	118 → 118
BenchmarkEngineLIndexHit	183 → 182	-0.3%	1 → 1
BenchmarkEngineLLen	159 → 159	+0.1%	0 → 0
BenchmarkEngineLPush	10158 → 11053	+8.8%	18 → 18
BenchmarkEngineLRange	448 → 461	+3.0%	11 → 11
BenchmarkEnginePut	1529 → 1522	-0.5%	18 → 18
BenchmarkEnginePutParallel	1403 → 1463	+4.3%	18 → 18
BenchmarkEngineSetAdd	9572 → 9548	-0.3%	4 → 4
BenchmarkEngineSetCard	164 → 165	+0.4%	0 → 0
BenchmarkEngineSetContainsHit	109 → 103	-5.0%	0 → 0
BenchmarkEngineSetContainsHitParallel	122 → 123	+1.1%	0 → 0
BenchmarkEngineSetContainsMiss	175 → 174	-0.3%	0 → 0
BenchmarkEngineSetMembers	5846 → 5761	-1.5%	66 → 66
BenchmarkEngineSetRemove	1104 → 1145	+3.7%	1 → 1
BenchmarkEngineTopKAdd	1458 → 1511	+3.6%	21 → 21
BenchmarkEngineTopKAddParallel	1765 → 1846	+4.6%	21 → 21
BenchmarkEngineTopKListHit	570 → 590	+3.7%	11 → 11
BenchmarkEngineTopKListHitParallel	644 → 689	+7.0%	11 → 11
BenchmarkEngineZAdd	10237 → 10279	+0.4%	8 → 8
BenchmarkEngineZCard	159 → 160	+0.6%	0 → 0
BenchmarkEngineZRange	195342 → 196486	+0.6%	1017 → 1017
BenchmarkEngineZRangeByScore	195969 → 197903	+1.0%	1021 → 1021
BenchmarkEngineZScoreHit	103 → 104	+1.1%	0 → 0
BenchmarkStoreBitSet	113 → 113	-0.4%	1 → 1
BenchmarkStoreCGetHit	73 → 73	+0.2%	0 → 0
BenchmarkStoreCIncr	101 → 100	-0.5%	1 → 1
BenchmarkStoreCMSIncr	121 → 121	+0.0%	0 → 0
BenchmarkStoreCMSIncrN	123 → 121	-1.2%	0 → 0
BenchmarkStoreGeoAdd	9598 → 9582	-0.2%	2 → 2
BenchmarkStoreGeoPosHit	88 → 91	+2.9%	0 → 0
BenchmarkStoreGeoPosMiss	87 → 87	+0.0%	0 → 0
BenchmarkStoreGetHit	290 → 296	+1.9%	2 → 2
BenchmarkStoreGetHitParallel	334 → 336	+0.5%	2 → 2
BenchmarkStoreHGetHit	92 → 97	+5.3%	1 → 1
BenchmarkStoreHGetMiss	85 → 85	+0.2%	0 → 0
BenchmarkStoreHLLAdd	100 → 100	+0.6%	0 → 0
BenchmarkStoreHSet	10576 → 10619	+0.4%	3 → 3
BenchmarkStoreJsonSet	26312 → 26327	+0.1%	207 → 207
BenchmarkStorePut	301 → 300	-0.4%	2 → 2
BenchmarkStorePutParallel	361 → 354	-2.1%	2 → 2
BenchmarkStoreSetAdd	9212 → 9233	+0.2%	2 → 2
BenchmarkStoreSetContainsHit	89 → 90	+1.0%	0 → 0
BenchmarkStoreSetContainsMiss	85 → 84	-0.8%	0 → 0
BenchmarkStoreTopKAdd	980 → 985	+0.5%	20 → 20
BenchmarkStoreZAdd	9540 → 9556	+0.2%	2 → 2
BenchmarkStoreZScoreHit	86 → 86	+0.5%	0 → 0
BenchmarkStoreZScoreMiss	88 → 86	-2.6%	0 → 0

_{Posted by CI bench job. Not a merge gate.}

[Code0987]

Closing — you asked to review locally, not commit/PR.