Resistance modules and rigs suffer from diminishing returns and additionally the stacking penalty.

Any module or rig affecting a resistance gives a benefit equal to the base value as a percentage of the difference between your current value and the maximum of 100%.

For example, if your armour has a base 40% resistance to EM damage and you fit a module that grants 50% to EM resistance, your actual gain is 50% of 60% = 30% (60% being the difference between your starting 40% and the maximum of 100%) giving you a new value of 70%.

So by definition of the formula you are correct in that 100% is impossible unless CCP create a ship with a base resist of 100% or a module that gives 100%.

The stacking penalty reduces the base benefit of multiple modules affecting the same attribute roughly as the chart below:

First module: 100% of stated value

Second module: 86.9% of stated value

Third module: 58.9% of stated value

Fourth module: 32.4% of stated value

Fifth module: 14.3% of stated value

Sixth module: 5.0% of stated value

Seventh module: 1.3% of stated value

Eighth module: 0.3% of stated value

So you get very little benefit from too many modules. As Psyrelle points out the number of modules necessary to get close to 100% is basically slots poorly spent. In my experience 90% is a more realistic and all round more practical figure to aim for.