feat: add possibility to ignore unsupported fuses

[kyrylo-hrechykhin]

This PR introduces possibility to ignore not supported electron fuses specified in fuses config. Requirement for that behaviour is described in issue #10 .

This PR also introduces small changes in tests that make possible to run those tests on supported platforms without failing.

[kyrylo-hrechykhin]

@MarshallOfSound , could you please review this PR?
It introduces quite useful functionality and additionally fixes some small issues in tests.

[kyrylo-hrechykhin]

@MarshallOfSound , @codebytere , could you please review this change?

[kyrylo-hrechykhin]

Oh, I did not expect that asking to re-review @codebytere will remove request for review of @MarshallOfSound .
@MarshallOfSound , could you please review this PR?

[MarshallOfSound]

I'm not a fan of having an option that just disables a valid error message. I think this should be solved on the user side by consuming getCurrentFuseWire and only flipping fuses conditionally that are in that wire.

Bloating the API surface of @electron/fuses doesn't feel right IMO

[kyrylo-hrechykhin]

@MarshallOfSound , thanks for the review.
I am absolutely fine for not adding additional parameter to the @electron/fuses API as long as requirement mentioned in the #10 met.

I noted in the initial issue that current logic where fuses are set takes into consideration unsupported fuses already and just does not set those. In case of flipping no longer supported fuses it just warns in the console instead of throwing an exception. From my point of view, this is inconsistency in terms of handling not supported fuses.

I'd suggest to replace throwing of exception with console.warn as in handling of no longer supported fuses. With this change API won't change and behavior will be consistent for all unsupported fuses. And the most important thing, user of @electron/fuses will not be responsible for handling of combinations of fuses on different electron versions.

Users will still have the possibility to break the build by themselves, comparing fuses they would like to set with those returned by getCurrentFuseWire.

What do you think about it?