Harden tip-count guards and remove fixed-size RF complement buffers for SL_MAX_TIPS resilience

[Copilot]

This change starts implementation of the SL_MAX_TIPS resilience work by tightening guard behavior and removing assumptions that depend on fixed compile-time split dimensions. It aligns RF paths with existing guard semantics and makes failure modes explicit when tip counts exceed supported bounds.

• Guard semantics made consistent across RF C++ entry points

  • cpp_robinson_foulds_distance() now enforces check_ntip() (previously missing).
  • check_ntip() now throws a runtime R error (Rcpp::stop) with the existing user-facing message instead of relying on assertion behavior.
  • Bound is explicitly computed as min(SL_MAX_TIPS, int16_t::max()), with a static_assert documenting SL_MAX_TIPS/int32 assumptions.

• Removed fixed-size RF complement stack arrays

  • Replaced:
    • splitbit b_complement[SL_MAX_SPLITS][SL_MAX_BINS]
  • With runtime-sized heap storage in:
    • robinson_foulds_distance()
    • robinson_foulds_info()
  • This avoids hard coupling to static dimensions and prevents oversized stack allocations as limits evolve.

• Centralized R-side tip-limit check in tree-distance utilities

  • Added internal helper in R/tree_distance_utilities.R:
    • .AssertNtipSupported()
    • .MaxSupportedTips (documented to stay aligned with C++ guard semantics)
  • Utility guard call sites now route through this helper.

• Focused regression coverage for boundary behavior

  • Added tests in tests/testthat/test-tree_distance_utilities.R for:
    • valid boundaries (1000, 32766, 32767)
    • reject boundary (32768)
    • direct RF C++ entry points
    • wrapper path (.SplitDistanceAllPairs) using the same guard message

void check_ntip(const double n) {
  static_assert(SL_MAX_TIPS <= std::numeric_limits<int32>::max(),
                "SL_MAX_TIPS must fit in int32");
  constexpr int32 max_supported_tips = std::min(
    int32(SL_MAX_TIPS), int32(std::numeric_limits<int16_t>::max())
  );
  if (n > max_supported_tips) {
    Rcpp::stop("This many tips are not (yet) supported.");
  }
}

Original prompt

Start implementation

The user has attached the following file paths as relevant context:

• .github\copilot-instructions.md
• AGENTS.md

[Chronological Review: The conversation began with the user requesting a comparison between PR #193 and their own branch, focusing on how both address the issue of making TreeDist resilient to changes in SL_MAX_TIPS. The user detailed the differences in implementation, strengths, and weaknesses of both approaches. Following this, the user requested to start the implementation based on the evaluation provided, indicating a shift from analysis to action.]

[Intent Mapping: The user explicitly requested a comparison of PR #193 against their branch, highlighting specific features and shortcomings. They then requested to "start implementation," indicating readiness to proceed with coding based on the evaluation.]

[Technical Inventory: The conversation involved technical concepts such as:

• PR Prepare for large-tree support #193: Focused on static array elimination, centralized R guard, C++ guard, user interrupts, and CI/memcheck improvements.
• User's branch: Comprehensive migration addressing int16 to int32 migration, overflow protection, and extensive testing for large trees.
• Tools mentioned: Rcpp for user interrupts and error handling, and valgrind for memory checking.]

[Code Archaeology: The user is currently working on tree_distances.cpp, which is central to the implementation of distance calculations in the TreeDist project. The conversation highlighted specific functions and checks that need to be modified or added.]

[Progress Assessment: The user has completed the analysis of both PR #193 and their branch, identifying key areas for improvement. They are now poised to implement changes based on the recommendations provided.]

[Context Validation: All critical information for continuation is captured, including the need to cherry-pick improvements from PR #193 into the user's branch and the specific changes to be made in tree_distances.cpp.]

[Recent Commands Analysis: The last agent command was the user requesting to "start implementation." There were no tool results from this command as it was a directive to begin coding. The immediate pre-summarization state involved the user preparing to implement the changes discussed in the evaluation of PR #193.]

1. Conversation Overview: - Primary Objectives: The user requested a comparison of PR #193 against their branch, stating, "Both address the same core problem — making TreeDist resilient to SL_MAX_TIPS being raised in TreeTools." They then requested to "start implementation" based on the evaluation. - Session Context: The conversation transitioned from a detailed analysis of the two approaches to a directive to begin implementing the recommended changes. - User Intent Evolution: The user moved from analysis and comparison to readiness for implementation, indicating a clear plan to enhance their branch.

2. Technical Foundation:

• PR Prepare for large-tree support #193: Focused on static array elimination, centralized R guard, and user interrupts.
• User's Branch: Comprehensive migration to support larger trees, including int32 migration and overflow protection.

3. Codebase Status:

• File Name: tree_distances.cpp
• Purpose: Central to distance calculations in TreeDist.
• Current State: The user is preparing to implement changes based on the evaluation of PR Prepare for large-tree support #193.
• Key Code Segments: Functions related to distance calculations and checks for SL_MAX_TIPS.
• Dependencies: Changes will impact how distances are calculated and how the system handles larger trees.

4. Problem Resolution:

• Issues Encountered: Concerns about misleading claims in PR Prepare for large-tree support #193 regarding support for trees larger than 32767 tips.
• Solutions Implemented: Recommendations to cherry-pick improvements from PR Prepare for large-tree support #193 into the user's branch.

5. Progress Tracking:

• Completed Tasks: Analysis of PR Prepare for large-tree support #193 and identification of areas for improvement.
• Partially Complete Work: Implementation of changes in tree_distances.cpp is pending.
• Validated Outcomes: The user has a clear plan for the next steps.

6. Active Work State:

• Current Focus: The user is ready to implement changes in tree_distances.cpp.
• Recent Context: The user has just requested to start implementation after a detailed evaluation of PR Prepare for large-tree support #193.
• Working Code: The user is focused on modifying tree_distances.cpp based on the evaluation.
• Immediate Context: The user is set to begin coding the recommended changes.

7. Recent Operations:

• Last Agent Commands: The user requested to "start implementation."
• Tool Results Summary: No tool results were generated as this was a directive to begin coding.
• Pre-Summary State: The user was preparing to implement changes in tree_distances.cpp based on the evaluation of PR Prepare for large-tree support #193.
• Operation Context: The user is focused on enhancing their branch by integrating the best aspects of PR Prepare for large-tree support #193.

8. Continuation Plan:

• Pending Task: Implement changes in tree_distances.cpp ...

Created from VS Code.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 95.84%. Comparing base (65d9a37) to head (dd78ff5).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #194      +/-   ##
==========================================
+ Coverage   95.78%   95.84%   +0.05%     
==========================================
  Files          57       57              
  Lines        5508     5514       +6     
==========================================
+ Hits         5276     5285       +9     
+ Misses        232      229       -3     

☔ 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.

[github-actions]

Performance benchmark results

Call	Status	Change	Time (ms)
ClusteringInfoDistance(tr200)	⚪ NSD	1.13%	15.3 → 15.1, 15.3
ClusteringInfoDistance(tr50)	⚪ NSD	0.99%	11.6 → 11.4, 11.5
LAPJV(test2000)	⚪ NSD	0.9%	97.1 → 96.5, 95.4
LAPJV(test40)	⚪ NSD	-0.11%	0.0176 → 0.0177, 0.0175
LAPJV(test400)	⚪ NSD	0.01%	3.19 → 3.2, 3.19
MutualClusteringInfo(tr200)	⚪ NSD	1.86%	22.1 → 22, 21.3
MutualClusteringInfo(tr50)	⚪ NSD	1.14%	23.9 → 23.9, 23.5
PathDist(postTrees)	⚪ NSD	1.44%	3.64 → 3.57, 3.6
PhylogeneticInfoDistance(tr200)	⚪ NSD	0.23%	235 → 234, 234
PhylogeneticInfoDistance(tr50)	⚪ NSD	0.92%	81.9 → 81.5, 81
RobinsonFoulds(tr200)	⚪ NSD	1.12%	2.91 → 2.88, 2.88
RobinsonFoulds(tr200)	⚪ NSD	1.1%	2.68 → 2.65, 2.64
RobinsonFoulds(tr50)	⚪ NSD	0.22%	4.45 → 4.46, 4.43