fix: regression introduced with magazine reload fix

[usagirei]

Purpose of change (The Why)

Went to look back at what broke the mag reloading in the first place, it was introduced with #4817

#6174 partially reintroduced the issue, but in the form of you could then potentially load unsupported ammo into an empty magazine already inside a gun (300 Blackout on a STANAG on an AR-15)

Describe the solution (The How)

This should solve that by further checking if the item being reloaded is a magazine, that is contained on a gun/mod that doesn't support said ammunition.

Describe alternatives you've considered

N/A

Testing

Vanilla:

• Spawn AR-15, Mags, 556 and 300blk ammo
• Can insert 300blk ammo on an empty mag
• Mag with 300blk can't be inserted in AR-15
• Insert empty mag on AR-15
• Can't load the empty mag with 300blk any longer
• Add underslug shotgun or M203
• Reloading still shows up shot/grenades as potential ammo to reload

Modded:

• Attach MWPE Six12 underslug shotgun (magazine fed)
• Loose shot and magazines can be loaded as usual

Additional context

Should there be another check if the parent item is a gun / gunmod, instead of just checking if the magazine has a parent_item()?
Since we don't have pockets on BN, i couldn't imagine a situation where that would not be the case.

Checklist

Mandatory

• I wrote the PR title in conventional commit format.
• I ran the code formatter.
• I linked any relevant issues using github keyword syntax like closes #1234 in Summary of the PR so it can be closed automatically.
• I've committed my changes to new branch that isn't main so it won't cause conflict when updating main branch later.
• I understand that, unless specified otherwise, my contributions will fall under the AGPL v3.0 and/or CC-BY-SA 4.0 licenses

[chaosvolt]

awawa