Burn-in/out implementation #19
Description
Added a burnin feature as an njit function in core.py.
Burnin lets us model the starts/ends of scan paths when they occur inside the domain of interest; previously, all scan paths started and ended outside the domain (RVE/SVE assumption).
Goal is to model stripe boundaries observed in coupons (a la Zeiss coupon contouring) and the associated stochastic LoF defect formation.
Some todos:
- add interface with cli/api so that burn-in can be turned on or off and user-controlled
- separate burnin/out times (currently symmetric)
- right now, burnin time t0 is normalized relative to the scan path times. ideally though, we'd want to add to the scan path time, i.e. renormalize with [-t0,1+t0] so that the melt pool reaches full geometry (attenuation factor 1) at local normalized t=0.