Refactor engine into focused files to improve maintainability

Split monolithic engine.go (776 lines) into specialized files:
- engine.go: Core orchestration (142 lines, 82% reduction)
- collection.go: Collection processing and management (445 lines)
- content.go: Content injection and extraction (152 lines)
- discovery.go: Element discovery and DOM traversal (85 lines)

Benefits:
- Single responsibility principle applied to each file
- Better code organization and navigation
- Improved testability of individual components
- Easier team development and code reviews
- Maintained full API compatibility with no breaking changes

internal/engine/discovery.go

@@ -0,0 +1,84 @@
+package engine
+
+import (
+	"golang.org/x/net/html"
+)
+
+// InsertrElement represents an insertr element found in HTML
+type InsertrElement struct {
+	Node *html.Node
+}
+
+// findEditableElements finds all editable elements (.insertr and .insertr-add)
+func (e *ContentEngine) findEditableElements(doc *html.Node) ([]InsertrElement, []CollectionElement) {
+	// Phase 1: Pure discovery
+	insertrElements, collectionElements, containers := e.discoverElements(doc)
+
+	// Phase 2: Container expansion (separate concern)
+	expandedElements := e.expandContainers(containers)
+	insertrElements = append(insertrElements, expandedElements...)
+
+	return insertrElements, collectionElements
+}
+
+// discoverElements performs pure element discovery without transformation
+func (e *ContentEngine) discoverElements(doc *html.Node) ([]InsertrElement, []CollectionElement, []*html.Node) {
+	var insertrElements []InsertrElement
+	var collectionElements []CollectionElement
+	var containersToTransform []*html.Node
+
+	// Walk the document and categorize elements
+	e.walkNodes(doc, func(n *html.Node) {
+		if n.Type == html.ElementNode {
+			if hasInsertrClass(n) {
+				if isContainer(n) {
+					// Container element - mark for transformation
+					containersToTransform = append(containersToTransform, n)
+				} else {
+					// Regular element - add directly
+					insertrElements = append(insertrElements, InsertrElement{
+						Node: n,
+					})
+				}
+			}
+			if e.hasInsertrAddClass(n) {
+				// Collection element - add directly (no container transformation for collections)
+				collectionElements = append(collectionElements, CollectionElement{
+					Node: n,
+				})
+			}
+		}
+	})
+
+	return insertrElements, collectionElements, containersToTransform
+}
+
+// expandContainers transforms container elements by removing .insertr from containers
+// and adding .insertr to their viable children
+func (e *ContentEngine) expandContainers(containers []*html.Node) []InsertrElement {
+	var expandedElements []InsertrElement
+
+	for _, container := range containers {
+		// Remove .insertr class from container
+		RemoveClass(container, "insertr")
+
+		// Find viable children and add .insertr class to them
+		viableChildren := FindViableChildren(container)
+		for _, child := range viableChildren {
+			AddClass(child, "insertr")
+			expandedElements = append(expandedElements, InsertrElement{
+				Node: child,
+			})
+		}
+	}
+
+	return expandedElements
+}
+
+// walkNodes walks through all nodes in the document
+func (e *ContentEngine) walkNodes(n *html.Node, fn func(*html.Node)) {
+	fn(n)
+	for c := n.FirstChild; c != nil; c = c.NextSibling {
+		e.walkNodes(c, fn)
+	}
+}