So, 'something' must be able to turn a photon of one wavelength into a photon of another wavelength, otherwise we'd only have 1 wavelength of light in our unive

So, 'something' must be able to turn a photon of one wavelength into a photon of another wavelength, otherwise we'd only have 1 wavelength of light in our universe. Moreover, if we assume conservation of energy, changing a photon w/ wavelength A into B leaves over some energy C, that again, assuming some mechanism of changing energy into light, a new photon could be emitted. So which assumption is wrong? (1) Light comes in many different wavelengths (2) Conservation of energy (3) Non-light energy can be changed into light energy (4) Light color follows the photon model for the relation between energy and wavelength. (or just do a simple google search: https://arxiv.org/abs/2009.03039 https://arxiv.org/abs/2009.03039 where it's sounds like it's been observed directly)