Refactor architecture: eliminate auto-discovery and consolidate packages

- Remove auto-discovery entirely (~450 lines)
  * Delete internal/content/discoverer.go
  * Simplify enhancer to single-phase processing
  * Remove duplicate container expansion logic

- Consolidate repository implementations
  * Move internal/content/client.go → internal/db/http_client.go
  * Group all repository implementations in db/ package

- Add file utilities to engine following Go stdlib patterns
  * Add engine.ProcessFile() and ProcessDirectory() methods
  * Engine now handles both content processing AND file operations

- Move site management to dedicated package
  * Move internal/content/site_manager.go → internal/sites/manager.go
  * Clear separation of site lifecycle from content processing

- Preserve container expansion (syntactic sugar)
  * .insertr on containers still auto-applies to viable children
  * Container detection logic consolidated in engine/utils.go

Result: Clean architecture with single source of truth for .insertr processing

internal/engine/file.go

@@ -0,0 +1,97 @@
+package engine
+
+import (
+	"fmt"
+	"os"
+	"path/filepath"
+	"strings"
+
+	"golang.org/x/net/html"
+)
+
+// ProcessFile processes a single HTML file (following Go stdlib pattern like os.ReadFile/WriteFile)
+func (e *ContentEngine) ProcessFile(inputPath, outputPath, siteID string, mode ProcessMode) error {
+	htmlContent, err := os.ReadFile(inputPath)
+	if err != nil {
+		return fmt.Errorf("reading file %s: %w", inputPath, err)
+	}
+
+	result, err := e.ProcessContent(ContentInput{
+		HTML:     htmlContent,
+		FilePath: filepath.Base(inputPath),
+		SiteID:   siteID,
+		Mode:     mode,
+	})
+	if err != nil {
+		return fmt.Errorf("processing content: %w", err)
+	}
+
+	return writeHTMLDocument(outputPath, result.Document)
+}
+
+// ProcessDirectory processes all HTML files in a directory (following filepath.Walk pattern)
+func (e *ContentEngine) ProcessDirectory(inputDir, outputDir, siteID string, mode ProcessMode) error {
+	// Create output directory if it doesn't exist
+	if err := os.MkdirAll(outputDir, 0755); err != nil {
+		return fmt.Errorf("failed to create output directory: %w", err)
+	}
+
+	return filepath.Walk(inputDir, func(path string, info os.FileInfo, err error) error {
+		if err != nil {
+			return err
+		}
+
+		// Calculate relative path for output
+		relPath, err := filepath.Rel(inputDir, path)
+		if err != nil {
+			return err
+		}
+		outputPath := filepath.Join(outputDir, relPath)
+
+		// Handle directories
+		if info.IsDir() {
+			return os.MkdirAll(outputPath, info.Mode())
+		}
+
+		// Process HTML files
+		if strings.HasSuffix(strings.ToLower(path), ".html") {
+			return e.ProcessFile(path, outputPath, siteID, mode)
+		}
+
+		// Copy non-HTML files as-is
+		return copyFile(path, outputPath)
+	})
+}
+
+// writeHTMLDocument writes an HTML document to a file
+func writeHTMLDocument(outputPath string, doc *html.Node) error {
+	// Create output directory
+	if err := os.MkdirAll(filepath.Dir(outputPath), 0755); err != nil {
+		return fmt.Errorf("creating output directory: %w", err)
+	}
+
+	file, err := os.Create(outputPath)
+	if err != nil {
+		return fmt.Errorf("creating output file: %w", err)
+	}
+	defer file.Close()
+
+	return html.Render(file, doc)
+}
+
+// copyFile copies a file from src to dst
+func copyFile(src, dst string) error {
+	// Create directory for destination
+	if err := os.MkdirAll(filepath.Dir(dst), 0755); err != nil {
+		return err
+	}
+
+	// Read source
+	data, err := os.ReadFile(src)
+	if err != nil {
+		return err
+	}
+
+	// Write destination
+	return os.WriteFile(dst, data, 0644)
+}