video/vp6: Init nihav decoder with encoded size, not display size

[SuchAFuriousDeath]

Description

The Vp6Decoder wrapper was passing header.disp_w/disp_h * 16 to VP56Decoder::init, but those are the displayed dimensions, not the encoded ones.

Fixes #23662

Not sure if we want a test is necessary for a change like this.

Testing

I'd like to wait for the swf in #23662, if it's not provided, I guess we could make a test from scratch.

Checklist

• I, a human, have self-reviewed this PR and fully understand the changes within.
• I have made or updated tests where possible.
• All of my commits are properly scoped, compile successfully, and pass all tests.
• This PR does not make sense to split up into smaller PRs.
• An LLM was involved in the authoring of this code.

[torokati44]

This is probably correct (at least it feels like it should be), and I'm a bit embarrassed... 🫣
I also wonder how it went unnoticed for so long...
I guess it's just extremely rare to have mb and disp sizes differ.
And now I'm also wondering whether they are used incorrectly somewhere else as well... 🤔

[tygyh]

Weird edge cases are a great reason to add tests. If it was missed once it can be missed again.

[SuchAFuriousDeath]

I wouldn't call this an edge case, it is wrong, it's just rare for itself to manifest.