I think this is a great discussion and one of those topics that generally hasn't been fully discussed on this forum yet. Granted, I have not actually done a search for this topic on the forum, so I could very well be wrong.
Once again, Skip pretty much covered it as far as my knowledge base extends. Generally I can only agree and perhaps offer my opinion by stating it slightly differently than him. Perhaps some of the true experts like Jeff Snyder will see this and offer a more in-depth or more accurate explanation.
I think a case like El Reno 2013 shows that all of these features (from mesocyclone to sub-tornadic suction vortex) exist on a continuum of spatiotemporal scales such that there may not always be a clear distinction between them - i.e., different circulations may obtain length and time scales more characteristic of other types of circulations (e.g., a tornado approaching the size of a mesocyclone). We, as scientists, have not yet sampled enough cases to be 100% confident that there is a meaningful difference between suction vortices, a main tornado, a tornado cyclone, and a ground-hugging & damaging "mid-level" mesocyclone. We just don't have the data for an empircal analysis, and as far as I'm aware (limited awareness) no one is running modeling studies focusing on the distinction between these phenomena.
I'm a reductionist, meaning I tend to simplify complex objects and ideas by reducing them to progressively more fundamental notions, eventually arriving at the root level explanation. In that regard, I think the AMS definition of a tornado (repeated below) is the best place to start. The word "violently" was added by me - although it does appear in the definition from other sources (like the NWS).
AMS glossary said:
A violently rotating column of air, in contact with the surface, pendant from a cumuliform cloud, and often visible as a funnel cloud and/or circulating debris/dust at the ground.
Granted, this definition remains subjective (how "violent" is "violently"? How "often" is "often"? Is that detail even important?), and that hinders using it as the base level. However, I think it offers enough objectivity to be useful. In my opinion, if there is a column of air (not necessarily perfectly perpendicular to the average Earth gravity force vector - i.e., "vertical") attached to a cloud that contains significant vertical motion and development on the convective or tornado scale that is rotating fast enough to produce winds that cause damage to common surface objects like trees, homes, street signs etc, then it's all the same thing - a tornado. Are there dynamically-based differences between suction vortices and mesocyclones? Most certainly. But is that significant in regards to classification? That I'm not so sure of. Both features can cause damage and are harmful to human beings (who always seem to be seeking ways to classify things). However, they are the result of different processes.
TLDR: IDK, I'm just making an educated guess.