add a function that converts SFD EBV to 'true' EBV

[segasai]

Hi,

To address issues raised here
desihub/desispec#1159

I've added a function that converts SFD EBV to 'true' EBV, in order to avoid having a 0.86 normalization number in multiple places.

(To avoid possible problems related to diving by .86 instead of multiplying, I explicitly take input EBVs rather than just return the normalization number.)

[coveralls]

Coverage decreased (-0.01%) to 68.032% when pulling c862240 on dust_sfd_function into 3ce49f0 on master.

[sbailey]

After the 0.1% run is over (next week?) @julienguy and @schlafly will standardize our desitutil.dust calculations and merge this 0.86 factor internally so that the user doesn't have to do it externally (which would likely result in some users not knowing to do so). Let's not merge this PR yet; they can decide if they want to build on on this PR or do it separately.

I do like the spirit of corrected_ebv = function(sfd_ebv) to avoid multiply vs. divide confusion.

[weaverba137]

This PR is superseded by #169.

[julienguy]

Closing because superseded by more recent PR.