add OptionT.liftF and update related docs and tests. Close issue #659

[lambdista]

No description provided.

[codecov-io]

Current coverage is 80.97%

Merging #667 into master will increase coverage by +0.01% as of 5ef2fd9

@@            master    #667   diff @@
======================================
  Files          159     159       
  Stmts         2122    2123     +1
  Branches        72      72       
  Methods          0       0       
======================================
+ Hit           1718    1719     +1
  Partial          0       0       
  Missed         404     404       

Review entire Coverage Diff as of 5ef2fd9

Powered by Codecov. Updated on successful CI builds.

[ceedubs]

This may not really be something to worry about, but I'm thinking that maybe this should use Some(_) instead of Option(_). The Option.apply method has logic to turn null into None. In Cats we are generally assuming that people aren't passing around null. But if for some reason they really want to lift a null value into the OptionT structure, I don't think we should prevent them from doing that here.

[lambdista]

I used Option(_) to prevent null and getting None instead of Some(null) in that case, but if you confirm we should not prevent them from doing that I'll change it to Some(_).

[ceedubs]

I think that it should be consistent with our Applicative[Option].pure implementation, which uses Some: https://github.com/non/cats/blob/master/core/src/main/scala/cats/std/option.scala#L14. So yeah, I think we should change this here. I don't think we should conflate the issues of lifting and de-nulling.

[lambdista]

I totally agree! I'll change it to Some(_). Thank you.

[ceedubs]

Thanks, @lambdista! In general, this looks great. I left a couple very minor comments.

[lambdista]

Thank you @ceedubs for your support!

[ceedubs]

👍 (assuming the build passes). Thanks again!

[lambdista]

Thank you! 👍

[non]

👍 Good work! Thanks! 🐈