Raise SL_MAX_TIPS from 2048 to 32768

[ms609]

• SL_MAX_BINS: 32 -> 512, giving SL_MAX_TIPS = 64 * 512 = 32768
• Introduce SL_STACK_BINS (32) and SL_STACK_SPLITS (2045) as stack allocation thresholds, keeping SplitList objects at ~548 KB (not bloated)
• Trees with > 2048 tips use heap allocation in SplitList (transparent to callers)
• Update splits_to_tree.cpp to use stack thresholds for its local arrays
• Lookup table cost: ~1 MB (lg2 + lg2_double_factorial + lg2_unrooted)

[github-actions]

Performance benchmark results

Call	Status	Change	Time (ms)
as.Splits(bigTrees)	⚪ NSD	1.15%	22.6 → 22.4, 22.4
as.Splits(someTrees)	⚪ NSD	0.06%	11.2 → 11.2, 11
Consensus(forest1k.888, check = FALSE)	🟣 ~same	-3.52%	94.7 → 98.1, 97.9
Consensus(forest201.80, check = FALSE)	⚪ NSD	-1.01%	3.95 → 3.98, 4
Consensus(forest21.260, 0.5, FALSE)	⚪ NSD	-1.36%	1.2 → 1.23, 1.21
Consensus(forest21.260)	⚪ NSD	-1.94%	1.22 → 1.25, 1.22
Consensus(forestMaj, 0.5, FALSE)	⚪ NSD	-0.47%	2.91 → 2.94, 2.9
DropTip(tr2000, 5)	⚪ NSD	0.1%	16.8 → 16.7, 16.8
DropTip(tr80, 5)	⚪ NSD	-0.08%	0.105 → 0.105, 0.105
DropTip(unlen2k, 5)	⚪ NSD	1.69%	0.215 → 0.218, 0.204
DropTip(unlen80, 5)	⚪ NSD	-1.22%	0.0401 → 0.0414, 0.0396
lapply(bigSplits, as.phylo)	⚪ NSD	0.56%	29.7 → 29.7, 29.2
lapply(someSplits, as.phylo)	⚪ NSD	-6.28%	13.6 → 16.6, 13.7
PathLengths(tr2000, full = TRUE)	⚪ NSD	-2.08%	16 → 16.5, 16.3
PathLengths(tr80, full = TRUE)	⚪ NSD	-2.74%	0.0976 → 0.102, 0.0985
PathLengths(tr80Unif, full = TRUE)	⚪ NSD	-3.64%	0.0992 → 0.105, 0.101
RootTree(tr2000, 5)	⚪ NSD	-3.75%	0.378 → 0.386, 0.397
RootTree(tr80, c("t3", "t36"))	⚪ NSD	1.05%	0.0707 → 0.0698, 0.07
RootTree(tr80, "t3")	⚪ NSD	-1.8%	0.0495 → 0.0498, 0.0508
RootTree(tr80, "t30")	⚪ NSD	-0.62%	0.0499 → 0.0502, 0.0503
RootTree(unlen2k, 5)	⚪ NSD	-0.39%	0.329 → 0.332, 0.327
RootTree(unlen80, c("t3", "t36"))	⚪ NSD	1.2%	0.065 → 0.0644, 0.0641
RootTree(unlen80, "t3")	⚪ NSD	0.08%	0.0431 → 0.0432, 0.043
RootTree(unlen80, "t30")	⚪ NSD	-0.74%	0.0434 → 0.0438, 0.0436
TreeDist::RobinsonFoulds(forest201.80)	⚪ NSD	0.58%	15.9 → 15.8, 15.8
TreeDist::RobinsonFoulds(forest21.888)	⚪ NSD	0.86%	3.28 → 3.26, 3.23
TreeTools:::path_lengths(tr80$edge, tr80$edge.length, FALSE)	⚪ NSD	-2.74%	0.0882 → 0.0926, 0.0886
TreeTools:::postorder_order(bal40)	⚪ NSD	-0.68%	0.00163 → 0.00165, 0.00164
TreeTools:::postorder_order(bal40k)	⚪ NSD	0.96%	0.543 → 0.537, 0.538
TreeTools:::postorder_order(dbal40)	⚪ NSD	0%	0.0017 → 0.00171, 0.00168
TreeTools:::postorder_order(dbal40k)	⚪ NSD	-3.07%	2.06 → 2.12, 2.14
TreeTools:::postorder_order(dpec40)	⚪ NSD	0%	0.00253 → 0.00253, 0.00253
TreeTools:::postorder_order(dpec40k)	⚪ NSD	-0.05%	3300 → 3300, 3300
TreeTools:::postorder_order(drnd80)	⚪ NSD	0.25%	0.00406 → 0.00407, 0.00403
TreeTools:::postorder_order(nbal40)	⚪ NSD	-0.5%	0.00201 → 0.00202, 0.00202
TreeTools:::postorder_order(nbal40k)	⚪ NSD	1.34%	2.2 → 2.16, 2.18
TreeTools:::postorder_order(npec40)	⚪ NSD	0.36%	0.00284 → 0.00284, 0.00281
TreeTools:::postorder_order(npec40k)	⚪ NSD	0.02%	3320 → 3320, 3320
TreeTools:::postorder_order(nrnd80)	⚪ NSD	0%	0.00461 → 0.00463, 0.00459
TreeTools:::postorder_order(pec40)	⚪ NSD	-0.01%	0.00164 → 0.00165, 0.00164
TreeTools:::postorder_order(pec40k)	⚪ NSD	-0.8%	0.43 → 0.431, 0.436
TreeTools:::postorder_order(rnd80)	⚪ NSD	0.95%	0.00211 → 0.00212, 0.00206

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 96.06%. Comparing base (3c6b286) to head (bbe4679).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #265   +/-   ##
=======================================
  Coverage   96.06%   96.06%           
=======================================
  Files          80       80           
  Lines        5889     5890    +1     
=======================================
+ Hits         5657     5658    +1     
  Misses        232      232           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.