In this case the grooves just reduce the contact area, don't use them. If they are already there fine.
Fortunately these will be thicker tubesheets.
A rough hole ID can be very helpful.
When they clean the tubesheet holes they will need to gauge every hole to find any that are too oversized or out of round.
These will either need to be abandoned or reamed out and sleeved.
This is actually done a lot in power plants.
Make the service provider supply references. And check with them.
You must build and test mock-ups to get the rolling torque correct. You will want to roll very near to the minimum that will work.
When you hydro and find leaks you can re-roll those just a little harder. But you have to be careful if you over-roll them you will have 7 tubes that leak (the original and all of its neighbors because you deformed the tubesheet).
You cannot use apparent wall thinning since it will be near zero and exceeding difficult to measure.
The whole trick is to never cause yielding in the tubesheet.
You will also need to roll as near the edges of tubesheet as you reliably can. I have seen them use 1/8" in from the front and 1/4" from the back.
I presume that this is a re-tube.
So you are going to be staking the Ti tubes between each support plate for vibration resistance?
I have seen Superferric SS tubes put into Muntz tubesheets (about 5x difference in yield strength), so your case isn't as difficult.
= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed
