BUG-125: Wrong Variable Appended in Convergence Tracking in _update_edge()

[ColtonPayne]

Severity: CRITICAL
Location: scripts/interpretation/interpretation.py:1631

Description:
In _update_edge(), when updating an IPL complement predicate in the if p2 == l: block, the code updates p1's bounds but incorrectly appends p2's bounds to the updated_bnds list used for convergence tracking.

# Lines 1617-1633 in _update_edge
if p2 == l:
    if p1 not in world.world:
        world.world[p1] = interval.closed(0, 1)
        # ... predicate_map update
    
    if atom_trace:
        _update_rule_trace(rule_trace_atoms, ..., world.world[p1], ...)
    
    # Update P1's bounds
    lower = max(world.world[p1].lower, 1 - world.world[p2].upper)
    upper = min(world.world[p1].upper, 1 - world.world[p2].lower)
    world.world[p1].set_lower_upper(lower, upper)
    world.world[p1].set_static(static)
    ip_update_cnt += 1
    
    # BUG: Appending P2 instead of P1!
    updated_bnds.append(world.world[p2])  # ← Line 1631: WRONG!
    # Should be: updated_bnds.append(world.world[p1])
    
    if store_interpretation_changes:
        rule_trace.append((numba.types.uint16(t_cnt), numba.types.uint16(fp_cnt), comp, p1, interval.closed(lower, upper)))

Comparison with _update_node (CORRECT):

# Lines 1511-1527 in _update_node
if p2 == l:
    # ... same setup code ...
    lower = max(world.world[p1].lower, 1 - world.world[p2].upper)
    upper = min(world.world[p1].upper, 1 - world.world[p2].lower)
    world.world[p1].set_lower_upper(lower, upper)
    world.world[p1].set_static(static)
    ip_update_cnt += 1
    
    updated_bnds.append(world.world[p1])  # ← Line 1525: CORRECT!

Impact:

• Convergence detection broken: The convergence delta calculation (lines 1643-1647) iterates through updated_bnds to compute max delta. Using p2's bound (which wasn't actually updated) instead of p1's bound (which was updated) produces incorrect deltas.
• Only affects edges with IPL constraints: Nodes are unaffected since _update_node has correct code.
• Silent failure: No error raised, convergence just doesn't work correctly.

Fix:

updated_bnds.append(world.world[p1])

[ColtonPayne]

Fixed with PR #102