Speculatively implement LWG-3636: formatter<T>::format should be const-qualified

[fsb4000]

Fixes DevCom-1662387

[AlexGuteniev]

When making changes per non-WP-status lwg issue, you must comment every occurence, mention "per LWG-XXX"

[CaseyCarter]

When making changes per non-WP-status lwg issue, you must comment every occurence, mention "per LWG-XXX"

Technically, we require commenting behavior that contradicts the Working Draft. All of these const additions technically don't contradict the Working Draft; it requires some expressions to be valid, they can be made so regardless of whether format is const-qualified.

If we were to start calling format on const program-defined formatters, that would be a different story.

[barcharcraz]

Looks good, but the mutable is not good very bad.

Unfortunately we need to just copy _Specs into format before getting the dynamic parameters.

It's not a huge deal as this is only for user-defined types, and if it's a performance problem we can keep the non-const version (assuming wg21 doesn't prohibit that)

[CaseyCarter]

LGTM. Added to my "Stuff to port" list.

[StephanTLavavej]

I'm mirroring this to the MSVC-internal repo - please notify me if any further changes are pushed. (I'll handle verifying that this works with the recent LLVM update.)

[StephanTLavavej]

Thanks for time-traveling from the future when this LWG issue has been accepted! ⌚ 😹 ✅