I realize that the multiple items in the subject line of the original post greatly narrows the population of potential respondents, but I hope that I also conveyed the fact that I'm interested in respondents that don't necessarily meet all the the items. I mean, if someone indicates that it works with CyanogenMod at all, that's still significant information to me, though not as much as if it's the same version, on the same smartphone and laptop environment.
As for trial versions, for various reasions, I've already conceded to myself that I won't be able to give an app a sufficiently thorough test within the trial period. My plan is to commit to a solution, suck up whatever time costs are associated with it, be it due to configuration issues or other causes. If it doesn't work out, recover and move on to the next candidate.
Some of the problems I've read with other solutions is that it seems to work for some time, then corrupted data is noticed some time down the road. This makes it challenging to identify an ideal working solution within a short trial period, especially if I don't use the phone alot and don't have time to focus a good chunk of time to testing. It also means that the impact of corruption great, since it will have happened for some time before I notice. I won't know how far to go back in history to roll back my Address Book (assuming that I diligently make regular backups, which I won't), and it also means I lose any new info since the backup.
This makes it all the more important to make the initial decision as optimally as possible, even though the assessment contains uncertainties and unknowns. It will certainly be a decision made with far from perfect information. I appreciate the anecdote you provded regarding cyanogenmod.