Verify how the wave parameter is defined

[Ig-dolci]

Verify and potentially refactor the way the wave control parameter is set.

[Ig-dolci]

Adding Ruben's comment about this issue:

Fazer a distinção entre c e initial_velocity_model é necessário, pode ser que algum modelo precise calcular o grad com o modelo estendido por causa da formulação (espero que não). A outra situação é que o modelo estendido deve ser atualizado ao se atualizar o modelo inicial. De fato, ao aplicar as BCs, c é o modelo estendido e não mais o original porque é este quem o solver chama

[Olender]

I just want to add that there are multiple considerations to be taken into account here:

• Memory usage,
• Debugging,
• singe definition.

We should define c to be used in the propagation in a single place. I've had trouble in the past with this (where c was set multiple times throughout the code), where a wrong c sometimes appeared during inversion or remeshing and it led to me having to undertake significant refactoring of the code

[Olender]

We should talk about this on the Wednesday meeting

[Olender]

Ideas from Wednesday meeting mostly by @Ig-dolci :

• Kill initial_velocity_model in wave and use only c there;
• Add initial_velocity_model to FWI object;
• Let initial velocity model use the new grid dictionary structure thats capable of loading segys;
• Project and assign it to c in FWI object;
• CHeck if the assign is keeping it immuatable.

[Olender]

I'll work on the first 3