fix(test): repair the flaky offline==online acceptance; make the tie-break truly random

TASK 2 - THE FLAKY ACCEPTANCE TEST, root-caused. It was NOT a live/offline
boundary race as suspected. The replay spawned gun 13 (TMSelect) while live has
EnableTmSelector = false and never does. The shared VirtualTracker ring is
ORDER-SENSITIVE, so gun 13's extra 4 bullets/tick shift the ring head and
permute the per-tick RESOLUTION ORDER of every other gun. The learning guns
append observations in resolution order, so their predictions shifted and
produced small hit deltas that moved between runs.
Evidence: the first KNN divergence was at rtick=174 with the SAME resolution
set merely reordered (live ft133,138,139,142,148,150,151 vs offline
ft150,151,133,138,139,142,148); after closing gun 13's ready gate offline the
live and offline KNN traces became BYTE-IDENTICAL (diff empty, 904/904 lines).
Fix: mirror the live rack in the replay. No tick exclusion, no tolerance
loosening. Stability: 5/5 consecutive runs now report 12/12 exact, each with
enemyDied=true - the death boundary is included, not excluded. The proof is
now real rather than a lucky run.

TASK 1 - the tie-break was not random. randomize() was only reached
incidentally through initTsetlinGun(), so a rack without Tsetlin had a fixed
rand() stream and ties always resolved the same way across process restarts.
Added seedSelectorRng() after gun construction, honouring GUN_SELECTOR_SEED.
Evidence: unseeded, 6 separate processes gave different pick sequences; with
GUN_SELECTOR_SEED=42, 3 processes gave identical sequences.

TASK 3 - PRUNING DOES NOT HELP; keep the full rack. 15 PAIRED runs per variant
vs DrussGT, 8 rounds, identical seeds:
  baseline             3238 shots  6.18%  (events 6.16%)  200 dmg/run
  Tsetlin disabled     3522 shots  5.76%  (events 5.71%)  197 dmg/run
  Tsetlin+Displace     3478 shots  5.46%  (events 5.37%)  183 dmg/run
Paired permutation tests: -0.34pp p=0.57 and -0.70pp p=0.21. Per-run
distributions completely overlap (baseline range [2.68, 10.00]; 15/15 and 14/15
runs inside it). A Crazy control showed no separation either. So removing the
measured-worst real performers is neutral-to-slightly-negative, and with
sd ~1.8pp a definitive claim either way would need far more runs.

CORRECTION TO A CLAIM I MADE: the 'virtual metric is INVERTED' finding does NOT
reproduce. Job-24 measured Spearman -0.374; this job measures +0.335 over the
same 13 guns with a different but equally defensible aggregation. Two opposite
signs means the correlation is NOT robustly negative - it is WEAK AND
SIGN-UNSTABLE. The honest statement is that virtual hit rate is a poor ranker,
not an inverted one. The docs assert the inversion and need correcting.

Also adds per-process GUN_STATS_PATH/GUN_SHOTLOG_PATH so concurrent A/B runs do
not clobber each other, and an env-gated GUN_RACK_DISABLE for rack A/Bs. All
default behaviour is unchanged when the env vars are unset.

common_libs/tests/acceptance_offline_vs_online.nim

@@ -78,7 +78,7 @@ proc main() =
 
   let online = lastOnlineRound(statsPath)
   let fx = loadFixture(recordPath)
-  let reports = replayFixture(fx, buildAllGunDrivers(), liveActual = true)
+  let reports = replayFixture(fx, buildAllGunDrivers(enableTmSelector = false), liveActual = true)
 
   # Map online stats by gun id.
   var onShots: array[14, int]

common_libs/tests/range_guns.nim

@@ -19,12 +19,21 @@ import guns/decay_gf
 import guns/knn_gun
 import guns/tm_selector
 
-proc buildAllGunDrivers*(seed = -1): seq[GunDriver] =
+proc buildAllGunDrivers*(seed = -1, enableTmSelector = true): seq[GunDriver] =
   ## seed >= 0 re-seeds the global RNG after constructing the stochastic guns
   ## (Tsetlin and the TM selector both call randomize() in their constructors),
   ## so their learning is reproducible for offline runs.
   ##
   ## Order matches ModularBot's gun ids exactly (TMSelect appended at 13).
+  ##
+  ## `enableTmSelector` must MIRROR the live rack. The shipped ModularBot has
+  ## `EnableTmSelector = false`, so the live loop never spawns gun-13 virtual
+  ## bullets. The replay's shared VirtualTracker ring is order-sensitive: extra
+  ## gun-13 spawns shift the ring head and permute the per-tick resolution ORDER
+  ## of every other gun, which scrambles the `obs` insertion order of the
+  ## learning guns (KNN, DecayGF) and makes the offline metric diverge from the
+  ## live one. Callers that mirror the shipped bot must pass false (the
+  ## acceptance test does); tests that specifically exercise TMSelect pass true.
   var tsetlin = initTsetlinGun()
   var tmSelector = initTmSelectorGun()
   if seed >= 0:
@@ -45,6 +54,10 @@ proc buildAllGunDrivers*(seed = -1): seq[GunDriver] =
     makeDriver("KNN",        initKNNGun()),
     makeDriver("TMSelect",   tmSelector),
   ]
+  if not enableTmSelector:
+    ## Same effect as the live `if EnableTmSelector` gate: never spawn gun 13.
+    ## Keep the slot so gun ids / report indices are unchanged.
+    result[13].readyCb = proc(): bool = false
 
 proc makeTsetlinDriver*(seed = -1): tuple[driver: GunDriver, gun: ref TsetlinGun] =
   ## Same as makeDriver("Tsetlin", ...) but keeps a handle to the concrete gun,