v0.1.3

Other

• automatic mode selection for QR-family and Han Xin build_text
• image sampler — camera detection for all 34 size variants
• image sampler — camera detection for all 32 rectangular sizes
• image samplers — camera detection via native fiducials

Added

• automatic mode selection: build_text for QR, Micro QR (new), rMQR and Han Xin
  now splits text into the cheapest mix of numeric / alphanumeric / byte segments
  via a shared cost-model dynamic program (segment::optimize_segments)