Expand generic factory to include relative and proportional support.

[anothersimulacrum]

Summary

SUMMARY: Infrastructure "Allow using relative and proportional with generic_factory"

Purpose of change

In the hope of eventually removing the assign() functions, and expanding support for generic factory (and continuing to support existing use cases), generic_factory needs to support relative and proportional.

Also, some enums needed to be converted to enum classes so the compiler wouldn't get confused and try to use them with proportional.

Fixes #43117

Describe the solution

To do this, do some template vodoo.
Create two templated structs, and take advantage of SFINAE (yeah, I had to look that up) with these to determine which template function to use.
Three basically discriminate between whether or not the type can specify value proportionally or relatively.
For proportionally, this is whether or not the type can be multiplied by a float, and for relative, this is whether or not the type can use the += operator with itself.

Also, add some error checking for when a proportional value is inappropriately specified.

Huge thanks to jbytheway for helping me use template vodoo to accomplish this.

Testing

Game loads without error. Bio-chitin armor has acid protection.

[jbytheway]

Maybe do the conversion to enum class as a separate PR first to make this easier to review?

[jbytheway]

Based on the gcc 8 compiler error it looks like bool is supporting proportional (and it's probably supporting relative too), which you probably don't want. Might need to explicitly blacklist that, e.g. by explicitly specializing supports_relative<bool> etc.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. Please do not 'bump' or comment on this issue unless you are actively working on it. Stale issues, and stale issues that are closed are still considered.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. Please do not 'bump' or comment on this issue unless you are actively working on it. Stale issues, and stale issues that are closed are still considered.

[anothersimulacrum]

Fairly sure everything here is working and is in a good enough state for this to be merged. So I think this is just waiting on 0.F.

[Wishbringer]

Should this have 0.F feature freeze tag?
This fixes ammo types that use copy-from and relative damage_type with armor_penetration.
Example ammo id "50_mk211" gets 0 penetration rather than +20 without this.
Seems like a fairly serious bug to have in a milestone release.

[anothersimulacrum]

As the person who introduced that bug, as far as I can tell, it's a display issue. As long as the damage is loaded using assign(), which it is right now, relative/proportional work correctly for it. You can test this using 9mm against an armored zombie if you're interested (and editing values).
I do intend to look into that and fix it before 0.F, but it's not a loading issue as far as I can tell.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. Please do not 'bump' or comment on this issue unless you are actively working on it. Stale issues, and stale issues that are closed are still considered.