faster matching of known services

[rade]

We hit this code a lot during rendering.

I have tested rendering performance against a recent Weave Cloud prod report. Run scope like this:

$ prog/scope --mode=app --weave=false --app.collector=file:/tmp/prod-on-dev-report.json

and then time the fetching of a topology with

$ time curl -so /dev/null http://localhost:4040/api/topology/<topology>

and then repeat for other topologies (NB: scope must be restarted in-between in order to defeat memoisation between requests).

Results (all times in seconds):

topology	master	branch
processes	0.14	0.11
containers	0.92	0.67
pods	0.74	0.48
services	0.72	0.45
hosts	0.76	0.47

So that's a nearly 40% improvement in some cases.

[2opremio]

Great catch @rade