Fill missing values for NMCP distributions

[bertozzivill]

• [x] Botswana 2014-- nvm, we don't model Botswana
• [x] Cote D'Ivoire: 2005/07/08 use interpolated years; 2012 look for PMI MOP
• [x] DRC 2005: Interpolate years (no docs found)
• [x] Gabon in general--can interpolate I guess, model overestimates anyway. EDIT: nvm, leave as zeros.
• [x] Mauritania 2007: Interpolate
• [x] Chad 2012: Set to 2013 value
• [x] Togo 2009: Set to 2020 value
• [x] Tanzania 2013-16 (filled w/ MOPs)