track the peak Allocator counters

[arvidn]

When compiled with the counters feature.
This is the more interesting metric when restore_checkpoint() is used.

---

Note

Low Risk
Adds new optional (feature-gated) metrics and threads them through the counters output without changing allocation limits or consensus logic; risk is limited to counter correctness and minor overhead when enabled.

Overview
When built with the counters feature, the allocator now tracks peak (max) allocated atoms, pairs, and heap bytes over the lifetime of a run (excluding ghost allocations) to better reflect memory pressure even when restore_checkpoint() frees allocations.

Allocator updates these max counters on atom/pair allocations and exposes max_* getters; run_program_with_counters() now returns these new fields, and tests were extended to assert the max counters are not reduced by restore_checkpoint().

^{Written by Cursor Bugbot for commit c68373b. This will update automatically on new commits. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.}

[coveralls-official]

Pull Request Test Coverage Report for Build 21900906247

Details

• 38 of 38 (100.0%) changed or added relevant lines in 2 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.05%) to 90.528%

---

Totals
Change from base Build 21864448712:	0.05%
Covered Lines:	6623
Relevant Lines:	7316

---

💛 - Coveralls