Change equations to match physics

[unalmis]

I would prefer to keep stuff like this as independent quantities. For stuff like g_rt and g_tr, those are the same both physically and numerically, independent of how we parameterize or discretize things.

Stuff like theta_PEST_tt and lambda_tt are actually different things physically, its just that they happen to the same numerically because we define theta_PEST = theta + lambda, but in theory other relations are possible and equally valid physically.

Originally posted by @f0uriest in #1175 (comment)

[unalmis]

In that case I have created the linked branch to reflect that. @f0uriest you confirm with a thumbs up / thumbs down whether you still want to create new compute functions that just do data["theta_PEST_z"] = data["lambda_z"]; return data ? This could be a couple thousand lines of code. Maybe also @ddudt since related to #1171.

[f0uriest]

Ideally yes, I'd like to keep them separate, but we don't have to do everything at once. The main point was to not remove existing things and replace them with aliases (and as we add new things try to keep to that convention).