tPAC issues

[Moo-Marc]

A few issues have come up related to phase-amplitude coupling. I'm making this issue to keep track of what needs our attention.

1. The recent code changes for tPAC https://github.com/brainstorm-tools/brainstorm3/pull/611 most likely need to be applied in related PAC processes: surrogate tPAC and inter-regional (IR) tPAC. (others?)
2. More generally, the 3 functions should be made consistent, ideally reuse the same code, the tPAC Compute function for example. Currently, IR does interpolation and averaging differently for example.
3. On top of fixes to the PAC averaging process https://github.com/brainstorm-tools/brainstorm3/pull/617, there's a need to clarify and/or correct what the averaging option does within the tPAC process itself. process_pac_dynamic: concatenates the "full" arrays (DynamicPAC field) which is already peak values in f_P, while averaging the "max across f_A" values (TF field). This is currently inconsistent with visualization (only displays first trial). process_pac_dynamic_sur2: no documentation in code or tutorials of what this does. Averaging is same as in process_pac_dynamic. process_pac_ir_dynamic: this one computes the average of DynamicPAC, unlike the other 2.
4. Code documentation is lacking.
5. IR doesn't properly deal with unconstrained sources for the "seed" side (but no error). And it uses the "seed" vertex only, which isn't clear. Probably a scout function should be used instead?