feat: implement comprehensive asset shipping for md-render command

Add automatic asset copying when rendering markdown to different output
directories with intelligent defaults and full user control.

Key Features:
- Environment variable support: MARKITECT_OUTPUT_DIR sets default output directory
- Smart defaults: auto-ship assets for directory output, disabled for file output
- CLI control flags: --ship-assets and --no-ship-assets for explicit control
- Timestamp-based copying: only copies when source newer than destination
- Path preservation: maintains relative directory structure in output
- Graceful error handling: missing assets logged as warnings, not failures

Technical Implementation:
- Enhanced asset discovery in markitect/assets/discovery.py with discover_assets_from_markdown()
- Added environment variable priority: CLI --output > MARKITECT_OUTPUT_DIR > input directory
- Comprehensive asset shipping logic with _ship_assets() function
- Directory vs file output detection for intelligent default behavior

Examples and Testing:
- Added image-assets example directory with 6 sample images and comprehensive README
- Created comprehensive TDD test suite with 10 tests covering all functionality
- Tests validate environment variables, CLI flags, asset discovery, shipping logic,
  timestamp handling, missing assets, path preservation, and default behaviors

Usage:
  markitect md-render file.md -o /output/dir/     # Auto-ships assets
  markitect md-render file.md --no-ship-assets   # Suppresses shipping
  MARKITECT_OUTPUT_DIR=/docs markitect md-render file.md  # Uses env var

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>

markitect/assets/discovery.py

@@ -223,6 +223,45 @@ class MarkdownScanner:
         return len(lines)
 
 
+def discover_assets_from_markdown(markdown_content: str, base_path: Path) -> List[AssetReference]:
+    """
+    Simple function to discover assets from markdown content for md-render.
+
+    Args:
+        markdown_content: The markdown content to scan
+        base_path: Base path for resolving relative asset paths
+
+    Returns:
+        List of AssetReference objects found in the markdown
+    """
+    scanner = MarkdownScanner()
+
+    # Create a temporary file to use the existing scan_file method
+    import tempfile
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.md', delete=False) as temp_file:
+        temp_file.write(markdown_content)
+        temp_path = Path(temp_file.name)
+
+    try:
+        references = scanner.scan_file(temp_path)
+        # Update the source_file to the actual base_path for relative resolution
+        for ref in references:
+            ref.source_file = base_path
+            # Resolve the asset path relative to base_path
+            if not ref.asset_path.startswith(('http:', 'https:', 'mailto:', 'data:')):
+                # Clean up relative path indicators
+                clean_path = ref.asset_path.lstrip('./')
+                resolved_path = base_path / clean_path
+                if resolved_path.exists():
+                    ref.resolved_path = resolved_path
+                else:
+                    ref.is_broken = True
+        return references
+    finally:
+        # Clean up temporary file
+        temp_path.unlink(missing_ok=True)
+
+
 class AssetDiscoveryEngine:
     """Main engine for asset discovery and analysis."""