I didn't follow why you think they're aren't enough supernovae. Let me throw out this thought using rough numbers.
A single particle in our tiny sun can experience about 4x1019 collisions in only a million years, based on a mean free path calculation.
There are about 1056 atoms in the sun. In a million years, our sun, by itself, can produce about 4x1075 collisions.
In one estimate, the stars in our galaxy contain about 1068 atoms. So if you allow for the same rate of collisions that our sun can produce, that gives about 4x1087 collisions for our galaxy as a whole in a million years.
If the universe is made of 400 billion galaxies, that gives about 1099 collisions without even taking into count the affects of a supernovae.
Now it's true that every collision won't result in fusion. But even if it's only one out of 1010, there are still 1089 collisions remaining without even considering supernovae.
This is enough collisions to replace all the atoms in the known universe (about 1080) with heavy nuclei in a million years assuming nothing more than stars like our sun.
This problem is really more complicated than this and I doubt I could come up with a good answer even in multiple sittings, but I thought I'd throw this out there.
Yes, and no.
To form a heavier nuclei, only those stars significantly (4 times sol's mass) heavier than our sun can create the pressure heavy enough to fuse more than than H+H -> .. -> He. And, in today's universe, the number of large stars is 1/1000 the number of brown dwarfs and and small stars. Interstellar dust,of course,also represents a "wasted" nova or supernova: that element was created, was successfully fused, but never got here.
Note: The lighter elements past the initial H1 and D2 and T3 He4 burning can (and do!) get formed in mainstream, ordinary suns our size and lighter, but the final products DON'T get ejected from the final white dwarf. Anything created in any mainstream star gets stuck in the center of a static cold star. Some, not all, mainstream stars can go nova, but they can't create heavy elements in that nova.
If a star does go nova, it adds to the general dust in space - important, but that dust faces the same "almost impossible" tiny chance of going in the right direction at the right time to get here as the supernova elements do.
It's the reason I focused my question on supernova's: they are (in today's universe, much less likely (by several factors of ten!) to occur than regular nova's.
For that matter, much of the matter created in a supernova has a good chance of being stuck in the black dwarf, black hole left in the center of the supernova. Again, for our planet's creation, that matter never got created in the first place, because it never got to our solar center's original dust cloud.
Any matter in another galaxy can't get here, so no other galaxy can be credited. Might be important to another being in another planet, but it doesn't contribute to our iron content. Nor Mar's core. Nor an asteroid's core.
Whether any matter created in any random part of our galaxy is important: I don't know of any mechanism where we can say "32 x 10^19 kg of iron were swept here from xxxx location so and so many billion years ago to form the cores of Mercury, Venus, Earth, and Mars - and another 5 x 10^19 kg of iron were swept here from that source, but lost into the sun and 120 x 10^19 kg of iron were formed in the right place, but came by before the sun's dust cloud was heavy enough to stop them."
You indicated that our galaxy has around 10^68 atoms.
We should question that: because the "rocky mass" in our little solar system, not counting any H or He, and not counting items out in the Oort cloud we can't find, has 10^50 heavy nuclei, and each of those represents only the material we know about that has undergone anywhere from 3 through 20 different fusion events. To fuse two carbon nuclei, for example, only takes one fusion event. One super high-energy collision as you pointed out. But, to get those two carbon nuclei in the right place to collide with other at the right temperature and pressure requires a whole series of previous collisions of exactly the right energy, direction, and pressure in the right star.
Thus, perhaps our "number of nuclei" count should be "number of fusions" represented BY the number of heavy elements we can measure.
If two carbon nuclei fuse in those 99/100 (995/1000 ?) stars too small to continue burning into a supernova, we don't care. As far as earth's core cares, they never existed.