Add ImTextureData-based texture management

[Igrium]

Added bindings for ImGui's new texture-management API: ImTextureData, ImTextureRect, ImTextureStatus, plus the relevant accessors.

Additionally:

• Wired up the default GL3 backend (ImGuiImplG13) to use the new dynamic font rendering
• Updated the example app to demonstrate this new flow

Notes:

• SDL backend has not been updated yet.
• Binaries only compiled for Linux (I do not own a Mac or Windows machine)

• Minor changes or tweaks (quality of life stuff)
• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

I maay have vibe-coded this lol

[SpaiR]

Thanks for the contribution - solid amount of work 👍

Two things I'd like addressed before merging:

1. ImTextureData/ImTextureRect are hand-written JNI rather than going through @BindingSource/@BindingField and buildSrc's generator, unlike every other struct in the binding (ImGuiStyle, ImFont, ImGuiViewport, ImFontConfig, ...). Manual JNI here should be the exception, not the default - reserved for parts the generator genuinely can't express (e.g. the direct-ByteBuffer pixel accessor). Could you rework it to go through the annotations like the rest of the codebase, and keep only what truly needs to be manual?

2. bin/libimgui-java64.so shouldn't be committed - natives are built by CI, not checked in with feature PRs. Could you drop that commit / revert bin/?

[Igrium]

Alright done. Also, the codegen decided to mess with TextEditor for some reason. I'm not sure what's up with that. Should I revert it?

[Igrium]

@SpaiR Is there anything else that needs to be done before it can be considered for merge?

[SpaiR]

Sorry, I'm struggling to find time for reviews right now.

On the PR itself — it still has blocking problems. The main one is the bin/libimgui-java64.so deletion together with the .gitignore entry, which you haven't fixed. That folder belongs to CI: CI's commit step uses a plain git add, so it would rebuild the Linux library on every run and commit it never. Drop both — CI replaces all three binaries after merge on its own.

The rest:

• Renderer_TextureMaxHeight docs say "width" — copy-paste, also in the generated copies.
• updateTexture hardcodes GL_RGBA where upstream asserts Format == RGBA32. An Alpha8 atlas would upload garbage with nothing to show for it; an early return is enough.
• createDeviceObjects used to unbind GL_PIXEL_UNPACK_BUFFER around the font upload. That guard didn't come along when uploads moved into renderDrawData, and the imports are still in the file.
• ImDrawData declares _GETTEXTURES_1 by hand. That name shape belongs to the generator — it's what @ReturnValue(isStatic = true) emits, e.g. _GETIO_1 in ImGui.java — and never appears in src/main/java. Rename it to TMP_TEXTURE_DATA to match getMonitors/getViewports in ImGuiPlatformIO. Otherwise it collides the day that method grows an annotation. The accessor itself is fine, that pattern is exactly right.
• Accessor.* star import → import GETTER directly, nothing else in the binding does this.
• Once RendererHasTextures is set, io.getFonts().build(), getTexDataAsRGBA32() and the deprecated createFontsTexture() all assert. Worth a line in the Javadoc and the release notes rather than being found at runtime.

Separately, and more importantly: I'm in the middle of fairly significant changes to the internal API myself, so I can't promise in the first place that this particular PR will be mergeable once I'm done — the approach to binding generation will most likely change quite a lot. So let it hang as is for now, and I'll ping you if and when it makes sense to update it.