As the grid moves away from physical inertia sources and loads, do you think it would be realistic to distribute a grid-wide signal separate from the actual line voltage which could assist non-rotating power sources to stay in sync or at least help reduce the chances of oscillation?
The easiest is probably radio or satellite broadcasts but the topology of the grid, which does change, would also have to be considered. Probably not an easy problem to solve simply?
The grid itself is the best source for that. What I think we will see instead is custom superconductance based sink/source units that help with local grid stabilization. Those are already being deployed and they work quite well absent mechanical solutions, but they are still expensive and their capacity is still limited. A really dumb (but probably quite effective) way of doing this could also be by simply hooking up massive but slow flywheels.
Both have the same effect. Good distribution of generation and consumption in a geographical sense is something we never really gave much consideration in the past, it wasn't rare at all to have one side of a geographic region to be 'mostly producers' and another to be 'mostly consumers' and where the two sat next to each other it was usually to accommodate some really large consumer (for instance, a paper mill or a steel or aluminum plant). That also allowed for co-generation which is far more efficient. I think we will see more of this as well, and incentives to allow EVs to be used as sinks during times of excess power availability.
Other options are HVDC interconnects between geographically distant regions or to use these to create micro grids, each of which would be less stable than a much larger one but it would serve to isolate problems if and when they occur.
Interesting detail: wind power, while theoretically rotating grid synchronized mass is increasingly uncoupled and powering the grid using inverters. This is for efficiency reasons, the rotors have a much wider range that way, and you then only use furling of the blades to protect the installation from overspeeding and maximum efficiency the rest of the time even if that means rotating at a different speed than what would sync with the grid. This is optional, if the machine is synchronized it will still produce power, but not quite as much because blades are more efficient at higher RPM running flatter than at lower RPM running coarse, though coarse they do have more torque. So by sticking an inverter in the middle you can basically electronically do MPPT for the windmill rather than doing that mechanically.
Over the life of an installation the cost of that inverter is more than paid back in extra power but it has the downside of not having the mechanical mass of the wind turbine rotor and blades as extra inertia. Win some, lose something else...