Add support for finite field isomorphisms and embeddings

[defeo]

As discussed at the Hecke/Nemo/Flint workshop, there is a plan to add support for lattices of compatibly embedded finite fields to Nemo/Flint. This issue tracks progress on the subject.

On the Flint side, we want to add the basic algorithms for computing and evaluating embeddings of finite fields. See https://arxiv.org/abs/1705.01221 for terminology and theoretical details. The higher level mechanics for compatible lattices will be coded in Nemo, as outlined in ...

We have an experimental branch, tracked in #351, where development is happening. We want to complete the following steps before pushing to trunk:

• [x] Write naive algorithms for computing embeddings;
• [x] Write algorithms to compute embedding matrices;
• [ ] Add a decent implementation of minimal polynomial computation for fq_* types;
• [ ] Write Allombert's algorithm as modified in https://arxiv.org/abs/1705.01221 (code can be ported from https://github.com/defeo/ffisom);
• [ ] Write helper functions to facilitate evaluating embeddings without computing matrices;
• [ ] Determine thresholds;
• [ ] Write docs.

Other useful algorithms that we could eventually add after the first part is merged:

• [ ] Other embedding algorithms, in particular Rains';
• [ ] Constructions for special families of finite fields (Lenstra-De Smit, De Feo-Doliskani-Schost, ...).

[wbhart]

Thanks for setting this up. I am away for about a week, so it will be a while before I get to look at all this. But I'm looking forward to ticking these off.