[Neural Link] Implement toJSON Serialization Protocol

Objective: Standardize how Neo.mjs instances are serialized for AI inspection (Neural Link). Currently, the Neural Link Client manually picks properties, which is brittle and unscalable. We will implement a toJSON() method protocol across the framework.

Core Concept: Every class decides its own JSON representation. JSON.stringify(instance) automatically calls toJSON().

Plan:

1. Phase 1: Foundation (core.Base, component.Base)

   • Implement toJSON() in core.Base: Export id, className, ntype.
   • Implement toJSON() in component.Base: Export key layout/visibility props (width, height, hidden, style, disabled).
   • Optimization: Ensure toJSON is non-recursive by default for complex object references to prevent circular dependency crashes and performance bottlenecks.

2. Phase 2: Common Widgets

   • Implement toJSON() for high-value components: Button (text, iconCls), Label (text), Container (layout).

3. Phase 3: Data & State

   • Implement toJSON() for data.Model (already exists? verify/standardize).
   • Implement toJSON() for data.Store (summary vs full dump).

4. Phase 4: Client Integration

   • Update NeuralLink_ComponentService (and others) to simply call instance.toJSON() instead of manually constructing objects.

Discussion Point:

• Tree vs. Detail: getComponentTree might need a "lightweight" serialization (just ID/ntype) vs getComponentProperty or queryComponent returning the "full" toJSON. We might need toJSON(detailLevel) or separate methods. For now, toJSON should be the "Detail View".

Value:

• Robustness: No more Client.mjs updates when a component adds a new config.
• Standardization: Uses native JS patterns.
• Scalability: Distributes the maintenance load to component authors.