`peak_spectra` function should not save the computed peak after every iteration

**Is your feature request related to a problem? Please describe.**

Currently, the `peak_spectra` function saves the intermediate data frame after every peak computation. This clogs up the process during the later peaks, since CSV writing gets slower as the data frame grows.

**Describe the solution you'd like**

Save at the end. If the process crashes, it's not a lot of work to re-run from the beginning.