Base readErrors on ddrescue map file instead of errors reported to stdout

[bitsgalore]

Currently Ipmlab's ddrescue module parses ddrescue's stdout for any reported errors, and the error count is reported as readErrors. Subsequently an non-equal value of readErrors sets the success flag to false.

In theory, this could lead to false positives in case of read errors that ddrescue subsequently recovers (e.g. using additional read passes).

An alternative approach would be to replace this error count with a check on the ddrescue map file. Format documented here:

https://www.gnu.org/software/ddrescue/manual/ddrescue_manual.html#Mapfile-structure

So basically we just need to read all data block lines and check the status character. For example:

# Mapfile. Created by GNU ddrescue version 1.23
# Command line: /usr/bin/ddrescue -b 512 -r4 -v /dev/sdb /mnt/p400/Diskette/Work/kb-f346d50a-ade4-11ed-a341-4370eb54a295/a5f7d186-adea-11ed-a341-4370eb54a295/a5f7d186-adea-11ed-a341-4370eb54a295.img /mnt/p400/Diskette/Work/kb-f346d50a-ade4-11ed-a341-4370eb54a295/a5f7d186-adea-11ed-a341-4370eb54a295/a5f7d186-adea-11ed-a341-4370eb54a295.map
# Start time:   2023-02-16 12:11:23
# Current time: 2023-02-16 12:12:54
# Finished
# current_pos  current_status  current_pass
0x00019200     +               4
#      pos        size  status
0x00000000  0x00019000  +
0x00019000  0x00001000  -
0x0001A000  0x0009A000  +

Here we have three blocks, with:

• Block 1: finished block ("+" char)
• Block 2: failed block bad-sector(s) ("-" char)
• Block 3: finished block ("+" char)

[bitsgalore]

First stab:

5abdb1d

Needs more testing + updated docs.

[bitsgalore]

Updated docs:

e62b4f4

All seems to work, but needs some more tests with real floppies.