The major reason for Gaussian peak shapes in gas phase or solution is Doppler broadening: since molecules are in motion, they "see" Doppler shifted wavelengths and this effect has a Gaussian distribution. There are also other broadening mechanisms that result in Gaussian distributions; if you are interested in more details, read the "Widths and Profiles of Spectral Lines" in Demtröder's Laser Spectroscopy textbook:

https://www.springer.com/de/book/9783642538582?msclkid=31642398659a16d9958243e55012ec85&utm_source=bing&utm_medium=cpc&utm_campaign=Springer.com%20%7C%20DE%20%7C%20Microsoft%20%7C%20Research%20%7C%20CPM%20%7C%20Shopping&utm_term=4584413737724039&utm_content=All%20Books

Another major factor in UV-Vis shaped are Franck-Condon-factors. Especially in nanoparticles you have numerous vibrations which you can coexcite with the electrons. Usually you will measure an envelope of various coexcited vibrations, but that's not generally Gaussian distributed. If the rotational constant of the molecule changes, you may observe effects such as so-called "band heads" or other special shapes. This is also described in Demtröder's book as well as a lot of other spectroscopy textbooks. If you are looking for a cheap alternative, you could check out "Molecules and Radiation" by Jeffrey Steinfeld.

I agree that Doppler broadening is important for gas phase spectra but I wouldn´t say Doppler is a major contribution to line-broadening in solution. In solution the major contributions to line broadening are known as homogeneous and inhomogeneous line broadening. Inhomogeneous lineshape is the easiest to tackle, arising from a static distribution of different frequencies each coming from a molecule with a different environment in solution, i.e. A solution is a distribution of molecules with different solvent environments each one contributing with a different frequency to the spectra. Normally this are randomly distributed and hence provide gaussian like spectra (take this with a grain of salt). Then there are the homogeneous contribution to line broadening. It is a little more difficult to picture: when a radiation field interacts with some material, a macroscopic polarization emerges as the sum of all oscillating electric dipoles of each molecule involved. This macroscopic polarization decays over time as a result of different processes, mainly the decay of the excited states and the loss of coherence between dipoles. Usually this decays lead to a Lorentzian lineshape, but slowly decaying processes can also lead to a Gaussian lineshape, although it is not good to over generalize.

You can read about this in most of nonlinear spectroscopy textbooks (since you can´t figure out the origin of linebroadening only by means of linear spectroscopy) . Mostly the book of Hamm and Zanni, and the one from Mukamel. You can also look at some lecture notes from Tokmakoff (tdqms.uchicago.edu/2d-spectroscopy-notes/). This paper might also help www.annualreviews.org/doi/pdf/10.1146/annurev.physchem.49.1.267

Hope this helps