Some corrections

[jsaintvanne]

Differents corrections :

• when you run it with few files and obtain no matching between samples, there is now an error message No results for these files... (https://github.com/workflow4metabolomics/metaMS/issues/37)
• change sapply into lapply because with 2 files there were an error because sapply and unlist didn't make a list and arguments hadn't the same length... It looks that works like this !
• revert new.annotations and annotations to have no error when the first result has 0 row (has mentionned here https://github.com/workflow4metabolomics/metaMS/issues/36)
• Add verification of matrix because when we have columns with the same length we obtain a matrix whereas we would like a list... (https://github.com/workflow4metabolomics/metaMS/issues/39)

[jsaintvanne]

@yguitton je retrouve une erreur lorsque j'enlève dans matcheSamples2Samples ce qu'on disait, donc la PR est bien comme ça ! :)