todo.txt

* In which gauges do we want to implement the interactions
-> I would advocate for Newtonian only. In this case we should:
   - Implement at test at the beginning of perturb_init
   - remove line 8053 in perturb_derivs: if (ppt->gauge == synchronous) dy[pv->index_pt_delta_nudm] = -metric_continuity;
      }


* check output in CAMB format if there is no nudm component
-> does it rise a segmentation fault?
-> is there a wrongly assigend nudm column title ?

* tests for the nudm implementation
-> does cdm and nudm (with zero interactions) give equvalent results
-> do 50/50 cdm and nudm (with zero interactions) give equivalent results

* ur species tests and verifications
-> does our code give equivalent results to my old implementation

* implement interactions in Boltzmann equations
-> Add interaction terms for ncdm l.8334 and on.
-> Add interaction terms for nudm l.8040 and on.

-> integration happening at l.4383
-> integration happening at l.6128
-> integration happening at l.7286
Calculate integrated C_chi in perturb_total_stress_energy?