Localization algebra

[jpoiret]

Hi,

This is an attempt to define the localization of an algebra at an arbitrary multiplicatively closed subset of it, compared with the current localization at a multiplicatively closed subset of the base ring. I've had many attempts, trying to use CT and the fact that algebras over a base ring are equivalent to rings under a base ring, but the definitional behavior was really hard to predict and things just didn't compute well. I chose instead to just extend the existing construction in the most straightforward way. I'm still trying to formulate a recursion principle using the universal property, so that we're able to prove things about the localization without resorting to its actual construction.

This depends on a rebased version of #931.

LMKWYT

[felixwellen]

Checking the library took 47min in the CI, we should investigate.

[felixwellen]

Both master and this branch take 27min on my machine, so it should be fine.

[felixwellen]

Great - thanks!