feat: allow identifierBase to be false

[lsvalina]

What

Added Disable Prerelease Identifier Base flag for creating prerelease identifiers without appending .0/.1
Improved Prerelease Identifier Base to use it when identifier is not set (1.0.0-{Identifier Base}) and adds description to cli help for -n flag

Why

Version 1.0.0-something is valid Semantic Version and its not necessary to have "build" number for identifiers that have only one/ first build, changes reflect that so if another prerelease is created with same identifier the "build" number will be append according Prerelease Identifier Base argument

References

Fixes#441
Improves #532

[wraithgar]

A "disable" boolean is often very confusing because it's turning something off when it's true.

We already have a flag to change this number, perhaps supporting false would suffice.

[wraithgar]

A very good start. Just some minor copy and code quality feedback.

[wraithgar]

@nlf what do you think about this? If they have said "increase the prerelease" but also asked for there not to be a prerelease base, do we default to 1 or should we throw?

[nlf]

my gut reaction is we should throw because they've asked for us to do something that doesn't make sense. going from no number to adding a number when we've been asked to not add numbers feels wrong

[wraithgar]

I agree, let's throw in this situation. No more guessing user intent!

[lsvalina]

just to clarify do we want to throw also in:

['1.2.0','prerelease','1.2.1-1',false,'',false],

?

[wraithgar]

Oooh good point. We sure do! "increase the prerelease with no identifier and no identifierBase" sure sounds impossible.

[lsvalina]

i updated it to throw on both cases

[wraithgar]

Can you add the tests from #550 to this PR?

[wraithgar]

Really useful new feature, thanks!