auto merge of #15418 : pnkfelix/rust/fsk-revise-VecPerParamSpace, r=pcwalton

In my informal measurements, this brings the peak memory usage when
building librustc from 1662M down to 1502M.  Since 1662 - 1502 = 160,
this may not recover the entirety of the observed memory regression
(250M) from PR #14604.  (However, according to my local measurements,
the regression when building librustc was more like 209M, so perhaps
this will still recover the lions share of the lost memory.)