Add naive sensor fusion support

Nobody has claimed this yet.

### Feature description

In general, the more information you have, the better your estimate will be. Alas, our bootstrap particle filter (BPF) workhorse can only reweight particles with one sensor model. We need mechanisms to aggregate multiple measurements, to perform sensor fusion. There are many such mechanisms depending on how you model your estimation problem, but I think we can cover quite some ground just assuming conditional independency and _multiplying_ 👀 sensor model likelihoods.

### Implementation considerations

Unclear as to whether this should a sensor model aggregate or a new `reweight` action 🤔. I'm slightly inclined towards the latter.

No files or tests are named. Start by locating the bootstrap particle filter's current single-sensor reweighting path and the sensor-model interface, then compare whether aggregation belongs in the sensor model or a new reweight action. Done means multiple measurements can be combined by multiplying their likelihoods under the conditional-independence assumption, with the chosen API covered by relevant tests.

Written by the indexing model from the issue text.