Add Likelihood Mode

[clementhelsens]

With this pull request:

• changed the default prior to "Uniform" instead of "DiscreteUniform" as it was causing problems with NUTS as it can only handle Uniform priors
• added an option to run the symbolic mode of the likelihood
  In a following PR I will also commit a piece of code that gets the mode of the likelihood from sampling, but it is less reliable in large dimensions

@francescorubbo , after merging this PR, I will move the code to the pyFBU organisation
https://github.com/pyFBU
to make it more standard.
I would need also the rights to deploy new releases through pypi, so if you could transfer them to me that would be great!

[francescorubbo]

@clementhelsens you are already owner, so you should have all the permissions.

[clementhelsens]

Thanks Francesco, I forgot to check pypi, and I just discovered I have an account! I confirm I have all the permissions. Le 1 nov. 2018 à 03:00, francescorubbo <notifications@github.com<mailto:notifications@github.com>> a écrit : @clementhelsens<https://github.com/clementhelsens> you are already owner, so you should have all the permissions. [screenshot from 2018-10-31 18-58-19]<https://user-images.githubusercontent.com/5140987/47828216-4976ca00-dd3f-11e8-9b49-b80364507fc7.png> — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#66 (comment)>, or mute the thread<https://github.com/notifications/unsubscribe-auth/AFBjDdHXTMMLfCv6pCOQbk7uJP4VZgjVks5uqlXbgaJpZM4YEQsX>.

[clementhelsens]

and @francescorubbo @gerbaudo I'm finalising the pyFBU document for arxiv submission. What institution should I use for you? maybe the best is to iterate via email, but I don't have an other address than your cern one, so maybe you could drop me an email?