[FEAT-Design/Planning]: Assess user story - Hot-starting failed runs

Title: Hot-starting failed runs
User Story:
As a modeler, I want continue failed simulations so that I don’t have to replicate already simulated outputs.
Notes:
The models operate on a statistical concept called Monte-Carlo simulation. The idea is that the overall simulation is consist of subsimulations (called iterations or life-cycles). Each iteration has elements sampled from stochastic distributions.

The models are not perfect, it is often the case that a specific iteration would trigger a model error and the overall simulation fails. If there are previously finished valid iterations, we want to continue on from where the last iteration finished rather than starting the computation from scratch.

I think this would fall into the category of optimization. So if the cost of restarting (in term of time and computing cost) is not very high, then this user story should not be as highly prioritized.

No files, tests, or entry points are named. First clarify how completed iterations are stored and identified, what should happen after an iteration fails, and how a resumed simulation is recognized as complete; done should be a decided, implementable scope for continuing valid prior iterations.