Edit to the above.
I've just reviewed my original idea. I had arranged the switching so that the meter was
always in cct. with series-connected resistors in shunt with it and the junctions of those resistors to the selector of the switch. Therefore if that selector failed to make contact with the shunt chain, the meter takes the full current. That's why I discounted it.
But in your arrangement, the series-connected shunts are
always across the meter with the switch selector connecting to the junctions of those shunt Rs - and that makes all the difference, of course!
The following extract from the book "Most Secret War' by R.V. Jones says it all:
Horace Darwin, the founder of the Cambridge Instrument Company, stated "Whenever you think that you have made a good design, try reversing the arrangement of some of the parts - it might work better".
'Nuff said!
Al.