"And we still need the lithium-7 to 99.99% purity (they are doing it in China, but the COLEX process is not legal here and we don't have an alternate). "
Do you though? For reference, Li7 is normally 92-5% of natural Lithium.
In the reaction process Li6 gets fissioned to tritium (which is a useful material for civil purposes) and you end up with free fluorine plus tritium gasses (easily extractable in the pump void space via sparging as is already done for xenon and helium removal) plus possibly some free uranium/thorium flouride materials which need to be separated out (bearing in mind that fissile fuel is only 1% of the salt load, the vast majority of fissioned Lithium doesn't alter the overall burn ratio). This can be dealt with by simple chemical separation and the result is that the lithium in the system is effectively self-purifying.
As you already have to cater to free flourine in the loops (it will be released by radiolysis whenever the fuel is cooled below freeze point) this doesn't really require extra materials engineering to handle it, although it does mean you need to have a vapor separator in the secondary loop to cater to (much lower levels of) lithium fissioning - but you need that anyway.
One of the good things about MSRs is that they do tend to lend themselves to "bucket chemistry" more readily than other designs. Using natural lithium will lower initial efficiency slightly, but you end up with a useful byproduct to go along with a number of other useful byproducts (all those noble gases have uses, even if you need to let the xenon sit around for a decade before selling it. Helium shortage? What Helium shortage?).
I'd be more worried about core design than lithium isotopes, and about beryllium toxicity from the current salt designs - which are a good reason alone to avoid LiFBe compounds.
The fact that a graphite core is eroded away over a 8-10 year period, requiring drain/rebuild is bad enough. The fact that you'll need to inert the atmosphere to prevent a fire inside the reactor vessel if you ever SCRAM the thing is far worse - the target should be _zero_ flammables - bear in mind that Chernoybl was so bad because it was a graphite moderator fire and the Windscale military reactor fire would have been worse than that if it wasn't for the insistence on "unnecessary" filtration devices fitted to the chimney (again, it was a graphite moderator fire)
On the bright side, the graphite problem appears to have been solved in a manner which also halves the physical size of the core. I can't find the cite at the moment, but at least one outfit has come up with a design based on zirconium ceramics which is erosion resistant.