Il have a look look at that rollover you mention .
There’s nothing lying in machine like diodes or bits of switches
Thinking back ,I think everything worked ,apart from maybe a switch for 1 solenoid
That doesn't tie up with an individual row or column wiring. To avoid confusion take a look at the matrix wiring diagram attached and let me know which switch numbers do work and which ones don't work out of the switches 16 - 45 on the diagram.
It appears that you have a problem with the switch matrix rows 5 and 3. Row 5 (switches 40 -45) is normally wired in green and goes to CN11 pin 6. Row 3 (switches 24 - 31) is normally wired in Yellow and goes to CN11 pin 4.
You need to check the continuity of these wires from the playfield back up to the relevant pins on CN11 and also check that they are not shorted to ground - power off the table and unplug CN11 to make these checks. To check for shorts to ground measure resistance from the pins on CN11 to the metalwork in the backbox. Compare the readings for pin 6 and pin4 to ground with the readings from the other row pins (3 and 5) for comparison.
If the wiring is all OK then it suggests a problem with the 3081 IC38 on the CPU board, but this should be OK based upon the tests Peter did on the board so I expect that you will find a wiring problem.
the fuseholders need replacing in due course but are not related to your switch problems.
The ground wire is a safety earth and should be repaired, but again is not related to your switch problems. I don't immediately recognise the wire clip you found, but hang on to it until you find out where it has come from!