update license

DESCRIPTION

@@ -8,7 +8,7 @@ Maintainer: Changwoo Lee <[email protected]>
 Description: Scalable methods for fitting Bayesian linear and generalized linear models in the presence of spatial exposure measurement error, represented as a multivariate normal prior distribution.
              These models typically arises from two-stage Bayesian analysis of environmental exposures and health outcomes, where predictions of covariate of interest (exposure) from a first-stage model and their uncertainty information are used in a second-stage regression model. 
              Related articles include Gryparis et al. (2009) (<https://doi.org/10.1093/biostatistics/kxn033>), Peng and Bell (2010) (<https://doi.org/10.1093/biostatistics/kxq017>), Chang, Peng, Dominici (2011) (<https://doi.org/10.1093/biostatistics/kxr002>), and Lee et al. (2024) (<https://arxiv.org/abs/2401.00634>). 
-License: MIT + file LICENSE
+License: GPL (>= 3)
 Encoding: UTF-8
 LazyData: true
 Roxygen: list(markdown = TRUE)