[lkdvos]

This PR adds StopWhenBelow(quantity, threshold), a generic criterion that stops once quantity(problem, algorithm, state) <= threshold, and indicates convergence.

• quantity is any function of (problem, algorithm, state), such as a gradient norm or a residual.
• StopWhenBelowState{T} records the last value and the iteration at which the threshold was reached, with T the floating point type of the threshold.
• get_reason and summary refer to the quantity by nameof(quantity), or as "value" for anonymous functions and callable objects.

gradient_norm(problem, algorithm, state) = norm(state.gradient)
StopWhenBelow(gradient_norm, 1e-8) | StopAfterIteration(100)

This change is non-breaking.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Files with missing lines	Coverage Δ
src/AlgorithmsInterface.jl	100.00% <ø> (ø)
src/stopping_criterion.jl	100.00% <100.00%> (ø)

... and 2 files with indirect coverage changes

🚀 New features to boost your workflow:

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

[lkdvos]

@jack-dunham or @mtfishman, it seems like this is a generically useful primitive, any comments or suggestions still welcome!

[mtfishman]

Looks good, thanks!