chore(benchmarks): tidy up benchmark workflows

[nlf]

drops some variables we don't consume anywhere, makes the event name something useful in the benchmarks repo, and updates the accept headers off of the old preview type

References

[npm-robot]

no statistically significant performance changes detected

timing results

app-large	clean	lock-only	cache-only	cache-only peer-deps	modules-only	no-lock	no-cache	no-modules	no-clean	no-clean audit
npm@8	45.351 ±3.00	26.656 ±0.23	15.929 ±0.06	18.469 ±1.01	2.695 ±0.01	2.685 ±0.00	2.188 ±0.05	10.601 ±0.07	2.145 ±0.00	2.870 ±0.02
#4145	43.824 ±0.34	27.148 ±0.00	16.054 ±0.11	18.715 ±1.29	2.720 ±0.01	2.736 ±0.00	2.199 ±0.08	10.691 ±0.05	2.160 ±0.01	3.116 ±0.32

app-medium	clean	lock-only	cache-only	cache-only peer-deps	modules-only	no-lock	no-cache	no-modules	no-clean	no-clean audit
npm@8	33.294 ±0.05	21.061 ±0.18	11.760 ±0.04	13.076 ±0.04	2.420 ±0.03	2.425 ±0.02	2.169 ±0.02	7.871 ±0.12	2.027 ±0.00	2.541 ±0.02
#4145	36.491 ±0.24	21.265 ±0.05	12.044 ±0.14	13.064 ±0.19	2.459 ±0.00	2.462 ±0.02	2.190 ±0.01	8.030 ±0.03	2.048 ±0.01	2.614 ±0.10