Smt

* add MFKPLS application (193, thanks m-meliani )
* add MFKPLSK application (198, thanks m-meliani )
* add parallel EGO with qEI criterion (202 , 199, 190, thanks EmileRouxSMB and rruusu )
* add notebook tutorial for EGO method
* fix kriging based surrogate bug (200 )
* fix full factorial sampling weights and clip options (197 )
* fix sklearn 0.22 cross_decomposition warning (196 )
* next releases > 0.5.x will likely **drop Python 2.7 support**

* fix EGO bugs (192)

* add <code>NestedLHS</code> sampling method in MFK application (187, thanks to Flo-code )
* fix Kriging based surrogates <code>predict_derivatives</code> when <code>poly</code> option is set to <code>"linear"</code> (188)
* fix VFM <code>predict_derivatives</code> and performance improvement (181)
* fix QP <code>predict_values</code> for multiple ouputs (ny>1) (180, 181)

* allow to pass initial DOE output (y) in EGO (185)
* doc fixes (179)

* rename _extensions_ as _applications_ (backwards-incompatible)
* add Efficient Global Optimization algorithm as a SMT application
* fix KRG, QP and LP surrogates when output dim > 1
* improve test coverage
* add SMT tutorial

* fix kriging predict_variances bug (158 , 159 )