v1.3.0

• changed names of two modules: "fitter" -> "optimize", "sampler" -> "sampling"
• revamped how optimization works; new procedure is streamlined for the user, and previous optimization code will no longer work. See /examples/optimize_example.py for a demonstration of the new framework
• bugfix for fitter.fit_model where negative parameter values are now handled correctly
• submodules are now imported immediately upon package import
• fitter.fit_model now warns you if you are not constraining the result and returns null errors rather than crash or produce garbage
• added some checks for data and hessian quality in error handling for fitter.fit_model
• default sun position changed to left-handed coordinate system for consistency
• fixed some imports around the package for consistency
• added sampler.perturb_value() for noise generation in quantity arrays