fix(xs-worker): respect !managerOptions.metered

[dckc]

goal: fixes #3074

This doesn't have an automated test for the main change.

Ideally I'd thread an instrumented version of spawn that records the args for the test to check. But spawn is imported from ambient authority in controller.js rather than being passed explicitly, so it's not obvious how to do this.

[dckc]

unrelated test failure in ci; for follow-up, see #3079

  message: 'newSwap doubleSwap threw: cannot coerce object to string',
  filename: 'test/unitTests/contracts/newSwap/test-newSwap-swap.js',

1 tests failed
F test/unitTests/contracts/newSwap/test-newSwap-swap.js newSwap doubleSwap

reproduced with

• yarn test:xs-unit test/unitTests/contracts/newSwap/test-newSwap-swap.js

failed to reproduce with

• yarn test:xs-unit test/unitTests/contracts/newSwap/test-newSwap-swap.js -m 'newSwap doubleSwap'
• yarn test:xs-unit --debug test/unitTests/contracts/newSwap/test-newSwap-swap.js

[dckc]

b646e7d tests that we can increment the counter and just not compare it against a threshold

[warner]

Yeah, it'd be nice to have that extra test, but it's not a big deal.

Hm, how messy would it be to build startXSnap with a spawn that recorded the arguments in a retrievable side-something and then called the real spawn? Probably too messy.

[dckc]

... how messy would it be to build startXSnap with a spawn that recorded the arguments in a retrievable side-something and then called the real spawn? Probably too messy.

On the contrary... a bit more ocap discipline doesn't look messy at all, to me. :) c885d23

[warner]

nice! hooray for ocaps