Fix variable bound violation in CPUFJ moves (#930)

## Issue

Authors:
  - Alice Boucher (https://github.com/aliceb-nv)

Approvers:
  - Rajesh Gandham (https://github.com/rg20)

URL: #930

Author: aliceb-nv

cpp/src/mip_heuristics/feasibility_jump/fj_cpu.cu

@@ -702,6 +702,21 @@ static void apply_move(fj_cpu_climber_t<i_t, f_t>& fj_cpu,
   raft::random::PCGenerator rng(fj_cpu.settings.seed + fj_cpu.iterations, 0, 0);
 
   cuopt_assert(var_idx < fj_cpu.view.pb.n_variables, "variable index out of bounds");
+  f_t old_val = fj_cpu.h_assignment[var_idx];
+  f_t new_val = old_val + delta;
+  if (is_integer_var<i_t, f_t>(fj_cpu, var_idx)) {
+    cuopt_assert(fj_cpu.view.pb.integer_equal(new_val, round(new_val)), "new_val is not integer");
+    new_val = round(new_val);
+  }
+  // clamp to var bounds
+  new_val = std::min(std::max(new_val, get_lower(fj_cpu.h_var_bounds[var_idx].get())),
+                     get_upper(fj_cpu.h_var_bounds[var_idx].get()));
+  delta   = new_val - old_val;
+  cuopt_assert(isfinite(new_val), "assignment is not finite");
+  cuopt_assert(isfinite(delta), "applied delta is not finite");
+  cuopt_assert((check_variable_within_bounds<i_t, f_t>(fj_cpu, var_idx, new_val)),
+               "assignment not within bounds");
+
   // Update the LHSs of all involved constraints.
   auto [offset_begin, offset_end] = reverse_range_for_var<i_t, f_t>(fj_cpu, var_idx);
 
@@ -761,17 +776,7 @@ static void apply_move(fj_cpu_climber_t<i_t, f_t>& fj_cpu,
   }
 
   // update the assignment and objective proper
-  f_t new_val = fj_cpu.h_assignment[var_idx] + delta;
-  if (is_integer_var<i_t, f_t>(fj_cpu, var_idx)) {
-    cuopt_assert(fj_cpu.view.pb.integer_equal(new_val, round(new_val)), "new_val is not integer");
-    new_val = round(new_val);
-  }
   fj_cpu.h_assignment[var_idx] = new_val;
-
-  cuopt_assert((check_variable_within_bounds<i_t, f_t>(fj_cpu, var_idx, new_val)),
-               "assignment not within bounds");
-  cuopt_assert(isfinite(new_val), "assignment is not finite");
-
   fj_cpu.h_incumbent_objective += fj_cpu.h_obj_coeffs[var_idx] * delta;
   if (fj_cpu.h_incumbent_objective < fj_cpu.h_best_objective &&
       fj_cpu.violated_constraints.empty()) {
@@ -786,11 +791,11 @@ static void apply_move(fj_cpu_climber_t<i_t, f_t>& fj_cpu,
       fj_cpu.iterations_since_best = 0;
       CUOPT_LOG_TRACE("%sCPUFJ: new best objective: %g",
                       fj_cpu.log_prefix.c_str(),
-                      fj_cpu.pb_ptr->get_user_obj_from_solver_obj(fj_cpu.h_best_objective));
+                      fj_cpu.pb_ptr->get_user_obj_from_solver_obj(fj_cpu.h_incumbent_objective));
       if (fj_cpu.improvement_callback) {
         double current_work_units = fj_cpu.work_units_elapsed.load(std::memory_order_acquire);
         fj_cpu.improvement_callback(
-          fj_cpu.h_best_objective, fj_cpu.h_assignment, current_work_units);
+          fj_cpu.h_incumbent_objective, fj_cpu.h_assignment, current_work_units);
       }
       fj_cpu.feasible_found = true;
     }
@@ -1021,6 +1026,13 @@ static void recompute_lhs(fj_cpu_climber_t<i_t, f_t>& fj_cpu)
   CPUFJ_NVTX_RANGE("CPUFJ::recompute_lhs");
   cuopt_assert(fj_cpu.h_lhs.size() == fj_cpu.view.pb.n_constraints, "h_lhs size mismatch");
 
+  // clamp to var bounds - defensive; apply_move should already have clamped appropriately
+  for (i_t var_idx = 0; var_idx < fj_cpu.view.pb.n_variables; ++var_idx) {
+    fj_cpu.h_assignment[var_idx] = std::min(
+      std::max(fj_cpu.h_assignment[var_idx].get(), get_lower(fj_cpu.h_var_bounds[var_idx].get())),
+      get_upper(fj_cpu.h_var_bounds[var_idx].get()));
+  }
+
   fj_cpu.violated_constraints.clear();
   fj_cpu.satisfied_constraints.clear();
   fj_cpu.total_violations = 0;