`lim_sup` -> `limn_sup`

[affeldt-aist]

analysis/theories/sequences.v

Line 2152 in 9fbd7cd

Definition lim_sup u := lim (sups u).

Would it be ok to rename lim_sup to limn_sup?
I am asking because there are "lim_sup" lemmas for realType in the pipeline
and it might be better to name these ones lim_sup (and not, say, limr_sup).
This also corresponds to the limn naming that the HB version of MathComp-Analysis is using.
Of course, a more unified mechanism would be better but that's another story for a bit later.