Refactor Estimator Accumulation to minimize warmup step cost

Currently in the batched drivers scalar and OperatorEstBase based estimators are alwars accumulated together.

In VMCBatched estimators are not accumulated during warmup.

In DMCBatched they must be because otherwise NaN's appear in scalar.dat output for the DMC section.

**Describe the solution you'd like**
TBD

**Describe alternatives you've considered**
A clear and concise description of any alternative solutions or features you've considered.

**Additional context**
Add any other context or screenshots about the feature request here.