When --rss is used, include an "RSS ratio" column

[eightbitraptor]

Looks like this when --rss is passed:

Average of last 10, non-warmup iters: 444ms
Total time spent benchmarking: 94s

interp: ruby 4.0.1 (2026-01-13 revision e04267a14b) +PRISM [arm64-darwin24]
yjit: ruby 4.0.1 (2026-01-13 revision e04267a14b) +YJIT +PRISM [arm64-darwin24]

------  -----------  ----------  ---------  ---------  ----------  ---------  ------------  -----------  ---------------
bench   interp (ms)  stddev (%)  RSS (MiB)  yjit (ms)  stddev (%)  RSS (MiB)  yjit 1st itr  interp/yjit  RSS interp/yjit
shipit  1010.2       7.1         240.4      444.7      3.8         239.1      2.448         2.272 (***)  1.005
------  -----------  ----------  ---------  ---------  ----------  ---------  ------------  -----------  ---------------

Legend:
- yjit 1st itr: ratio of interp/yjit time for the first benchmarking iteration.
- interp/yjit: ratio of interp/yjit time. Higher is better for yjit. Above 1 represents a speedup.
- RSS interp/yjit: ratio of interp/yjit RSS. Higher is better for yjit. Above 1 means lower memory usage.
- ***: p < 0.001, **: p < 0.01, *: p < 0.05 (Welch's t-test)

We don't do any confidence calculations here because we only measure the heap once per benchmark run per Ruby variation. We do GC and compact the heap first before taking a measurement.