Add 3 equality-constrained, regularized problems

I am interested in solving
```math
\min_x f(x) + h(x) \quad \text{s.t} \ c(x) = 0,
```
where every function can be non-convex and $$f$$ and $$c$$ are smooth.
I wish to implement

- [ ] The convex and non-convex Basis-Pursuit problems: $$min_x \lVert x \rVert_1$$ and $$min_x \lVert x \rVert_0$$ subject to $$Ax = b$$ : #89.
- [ ] The [Low rank matrix completion problem](en.wikipedia.org/wiki/Matrix_completion)
- [ ] The FH problem; this problem can be reformulated as $$\min_x \int |u(x,t) - u_{obs}(t)|^2 dt + \lVert x \rVert_0 $$ subject to $$du/dt = F(t,u,x)$$, we can use `PDENLPModels.jl` to model the smooth part of this problem.