"the conrod cap will always stretch more than the bolt, "filling up" any gap that will form"
I take it you mean this will occur during dynamic loading?
If so, it is certainly not true. Both components change in length during dynamic loading must be equal, until the joint separates.
You also seem to be implying that the bolt will be stiffer than the cap, again definetely not the case. If this were true, the bolt would be taking the majority of the alternating load.
You don't need a finite element model to calculate the axial stiffness of the con-rod bolted joint, just a simple hand calc.
There is absolutely no need to be investigating separation of the joint, the two bolts will be excerting a clamping force of much more than double the maximum inertia loading. I would put money on the fact that one of the bolts could provide enough clamping force.
Why don't you perform a calculation to eastimate the axial clamping force an M8 or M10 bolt produces when torqued up to a generic value or a percentage of yield. You would at least then have an appreciation of the joint clamping force and be clear in your own mind that it is not an issue.
If you want I can do a random calculation for an M8 or whatever size bolt's axial force.
Or you can attempt it yourself which would be more beneficial.