fix(log): pass in logger to external modules

[wraithgar]

Most of these module use npm-registry-fetch under the hood, which will
log things like the npm-notice header if seen. Currently we aren't
passing in a logger to them, which means that log message is never seen,
among any other logged messages those modules may need to make.

I added tests where I could. Some tests were in a state where the entire
libnpm* module was an empty mocked function, so asserting that it got
passed a log attribute was onerous.

[npm-robot]

found 20 benchmarks with statistically significant performance improvements

• app-large: clean, lock-only, cache-only, cache-only:peer-deps, modules-only, no-lock, no-cache, no-modules, no-clean, no-clean:audit
• app-medium: clean, lock-only, cache-only, cache-only:peer-deps, modules-only, no-lock, no-cache, no-modules, no-clean, no-clean:audit

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	56.306 ±0.53	36.000 ±0.18	22.488 ±0.09	25.277 ±0.56	3.635 ±0.04	3.659 ±0.14	2.914 ±0.05	14.984 ±0.00	2.881 ±0.03	4.079 ±0.09
#4359	0.479 ±0.00	0.522 ±0.03	0.482 ±0.00	0.481 ±0.01	0.485 ±0.01	0.486 ±0.03	0.490 ±0.00	0.517 ±0.02	0.492 ±0.00	0.486 ±0.01

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	43.522 ±2.12	27.296 ±0.00	16.257 ±0.08	17.641 ±0.02	3.356 ±0.01	3.297 ±0.08	2.947 ±0.08	11.051 ±0.30	2.763 ±0.03	3.631 ±0.10
#4359	0.483 ±0.01	0.470 ±0.01	0.462 ±0.00	0.504 ±0.00	0.504 ±0.03	0.505 ±0.01	0.469 ±0.00	0.491 ±0.01	0.480 ±0.01	0.495 ±0.01