Improve accuracy measure

[gilbertocamara]

Describe the requested improvement
The sits functions used for accuracy assessment, and more specifically sits_sampling_design() and sits_stratified_sampling() are based on the work of Olofssion et al (2014) which support the case where the strata match the map classes. However, there are other situations where one needs a more general approach, e.g., when comparing the output of different classifications. In this case, we need to include in sits the recommendations set forth by the following papers:

Stehman, Stephen V. 2014. ‘Estimating Area and Map Accuracy for Stratified Random Sampling When the Strata Are Different from the Map Classes’. International Journal of Remote Sensing 35 (13): 4923–39. https://doi.org/10.1080/01431161.2014.930207.

Stehman, Stephen V., and Giles M. Foody. 2019. ‘Key Issues in Rigorous Accuracy Assessment of Land Cover Products’. Remote Sensing of Environment 231 (September):111199. https://doi.org/10.1016/j.rse.2019.05.018.

Associated sits API function
sits_sampling_design() and sits_stratified_sampling()