For example, if I chose my inductance to be 3.3uH, my ripple current would be ~210mA. This would be ~70% of my maximum output current, and that value is way outside the suggested range of 20% - 40%.

This is what the data sheet says: -

It seems as though you are confusing maximum inductor current with maximum output current. Of course your ripple current can be high compared to your load current; it wouldn't be a good regulator if this wasn't true. The data sheet mentions maximum inductor current and, that is a rating of the inductor and the chosen maximum load current used.

If you need any further convincing, look at page 1: -

As you can see, that particular design works from a load current of 400 mA all the way down to 1 mA and, the only reason they stopped showing lower values is because the efficiency is pants at this low level.

Sure, if your load current maximum is only 300 mA it won't be a problem.

For example, if I chose my inductance to be 3.3uH, my ripple current would be ~210mA. This would be ~70% of my maximum output current, and that value is way outside the suggested range of 20% - 40%.

This is what the data sheet says: -

It seems as though you are confusing maximum inductor current with maximum desired output current. Of course your ripple current can be high compared to your load current; it wouldn't be a good regulator if this wasn't true. The data sheet mentions maximum inductor current and, that is both a rating of the inductor and the worst-case maximum load current used. So, if the design could deliver (say) 500 mA then your inductor needs to be appropriately rated (even if your expected full load current is smaller).

If you need any further convincing, look at page 1: -

As you can see, that particular design works from a load current of 400 mA all the way down to 1 mA and, the only reason they stopped showing lower values is because the efficiency is rubbish (naturally) at this low output level.

Sure, if your load current maximum is only 300 mA it won't be a problem.