bufnorm: Refactor and fix incremental bufNormalize

This fixes some edge cases the previous version didn't handle properly
by simplifying the logic of determining directly driven wires and
representatives to use as buffer inputs.

tests/various/bufnorm_opt_clean.ys

@@ -0,0 +1,154 @@
+
+read_aiger <<EOF
+aag 124 11 9 2 104
+2
+4
+6
+8
+10
+12
+14
+16
+18
+20
+22
+24 42
+26 57
+28 58
+30 68
+32 86
+34 113
+36 114
+38 105
+40 133
+228
+248
+42 26 5
+44 35 33
+46 36 3
+48 46 44
+50 36 31
+52 28 25
+54 53 50
+56 55 49
+58 38 25
+60 40 28
+62 60 25
+64 37 30
+66 64 24
+68 66 62
+70 29 7
+72 40 37
+74 7 2
+76 74 73
+78 76 12
+80 78 71
+82 76 13
+84 83 14
+86 84 80
+88 34 31
+90 89 58
+92 40 37
+94 39 26
+96 94 93
+98 96 91
+100 98 14
+102 39 6
+104 40 27
+106 104 103
+108 107 55
+110 109 98
+112 110 100
+114 24 10
+116 26 8
+118 7 5
+120 119 117
+122 5 3
+124 27 25
+126 125 123
+128 126 120
+130 129 83
+132 131 111
+134 48 12
+136 121 48
+138 137 135
+140 139 130
+142 138 81
+144 114 91
+146 145 57
+148 38 29
+150 148 25
+152 151 48
+154 153 69
+156 155 147
+158 157 143
+160 159 140
+162 147 81
+164 163 86
+166 165 21
+168 167 161
+170 169 22
+172 113 19
+174 133 113
+176 174 172
+178 108 83
+180 179 155
+182 181 19
+184 182 140
+186 185 177
+188 112 21
+190 189 176
+192 146 140
+194 192 164
+196 112 86
+198 196 165
+200 198 194
+202 181 156
+204 145 83
+206 205 17
+208 178 138
+210 209 206
+212 140 132
+214 212 210
+216 214 202
+218 194 161
+220 218 216
+222 220 201
+224 222 190
+226 224 186
+228 226 170
+230 167 157
+232 182 172
+234 193 21
+236 235 232
+238 202 197
+240 238 185
+242 238 188
+244 242 240
+246 244 237
+248 246 231
+c
+aigfuzz -s -1 1145242652
+seed 1145242652
+fuzzer layers
+depth 9
+width 19
+input fraction 13%
+latch fraction 84%
+lower fraction 20%
+monotonicity -1
+and closure
+EOF
+
+bufnorm -update
+
+delete -output o:* %R1
+opt_clean
+bufnorm -update
+
+delete -output o:* %R1
+tee -q debug opt_clean
+bufnorm -update
+
+bufnorm -reset
+opt_clean