Decouple low pass filters of eligibility trace and firing rate

[akorgor]

This PR undoes commit 2783bd6 and aims to fix learning performance issues for the case of kappa=0.

[JesusEV]

@akorgor Agnes, there is one technical aspect that is missing. The ideas shown in the lines, eprop_archiving_node.cpp:213 and, for example, eprop_iaf.cpp:427 are based on the factorization described in the attached document.
firing_rate_reg.pdf

[JesusEV]

@akorgor The sentence "According to Bellec et al. (2020), equation (2) suggests" would be better read as "According to equation (2) from Bellec et al. (2020)".

[JesusEV]

@akorgor I can address the missing steps by creating a new branch from this one. Once I'm done, you can merge it back into this branch. Let me know if that works for you.

[akorgor]

@akorgor I can address the missing steps by creating a new branch from this one. Once I'm done, you can merge it back into this branch. Let me know if that works for you.

Ah true, very good point. Please go ahead with the fix!