Uh, they could as a start include the theoretical predictions for all spectral lines?

Instead they take the subset of lines that have ended up in one (out of several) databases on spectral lines and, without any real motivation, declare this to be a complete sample.

Finally, putting the main result (Figure 2) before the section on data collection ("Experimental Section") is just rude.

Also, is it bad that the conflict of interest was disclosed?

In the context of what this paper brings to the reader, this should not have been published. If the paper at least did the chores and proposed some scientifically valuable hypothesis, then it could be OK to publish. But. If one guy had the original idea, worked his way through the required steps and the other discussed with him a little, he should have published himself, possibly with acknowledgement of the others. There is no work for 4 people here. The references are ridiculously long and nobody is going to read them all in order to "get" this paper.

Clearly, this is a product of the cultural, societal and economic pressure to "get published" often, quality or value be damned. My cousin who knows a little about science or academia once told me scientist should publish at least once a month, otherwise they do too little work. Obviously, the academia agrees.