annotate: build format string separately from annotation data
This prepares for porting to generic templater API.
Note that we cannot use '%*s' to pad white spaces because it doesn't take
into account character widths, as described in
4f5a6df2af92.
% ancestors of [] and not of [1]
[]
% ancestors of [] and not of []
[]
% ancestors of [12] and not of []
[0, 1, 2, 4, 6, 7, 9, 12]
% ancestors of [0] and not of [0]
[]
% ancestors of [4, 5, 6] and not of [6, 5, 4]
[]
% ancestors of [-1] and not of [12]
[]
% ancestors of [12] and not of [-1]
[0, 1, 2, 4, 6, 7, 9, 12]
% ancestors of [12] and not of [9]
[12]
% ancestors of [9] and not of [12]
[]
% ancestors of [12, 9] and not of [7]
[6, 9, 12]
% ancestors of [7, 6] and not of [12]
[]
% ancestors of [10] and not of [11, 12]
[5, 10]
% ancestors of [11] and not of [10]
[3, 7, 11]
% ancestors of [11] and not of [10, 12]
[3, 11]
% ancestors of [12] and not of [10]
[6, 7, 9, 12]
% ancestors of [12] and not of [11]
[6, 9, 12]
% ancestors of [10, 11, 12] and not of [13]
[0, 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 12]
% ancestors of [13] and not of [10, 11, 12]
[8, 13]
% lazy ancestor set for [], stoprev = 0, inclusive = False
[]
% lazy ancestor set for [11, 13], stoprev = 0, inclusive = False
[7, 8, 3, 4, 1, 0]
% lazy ancestor set for [1, 3], stoprev = 0, inclusive = False
[1, 0]
% lazy ancestor set for [11, 13], stoprev = 0, inclusive = True
[11, 13, 7, 8, 3, 4, 1, 0]
% lazy ancestor set for [11, 13], stoprev = 6, inclusive = False
[7, 8]
% lazy ancestor set for [11, 13], stoprev = 6, inclusive = True
[11, 13, 7, 8]