fix todo lint

Author: Wumpf

crates/re_renderer/src/allocator/cpu_write_gpu_read_belt.rs

@@ -55,18 +55,12 @@ pub struct CpuWriteGpuReadBuffer<T: bytemuck::Pod + Send + Sync> {
     _type: std::marker::PhantomData<T>,
 }
 
-// BufferViewMut is not Send/Sync right now because `BufferMappedRange` trait does not implement Send/Sync.
-// However, on native it is always Send/Sync, see https://github.com/gfx-rs/wgpu/issues/3795#issuecomment-1561189339
-// and on web we're single threaded.
-//
-// https://github.com/gfx-rs/wgpu/issues/3795 is resolved, we force Send+Sync here.
-
 #[allow(unsafe_code)]
-// SAFETY: TODO(https://github.com/gfx-rs/wgpu/pull/4818): Upstream wgpu allows `wgpu::BufferViewMut` to be Send.
+// SAFETY: TODO(gfx-rs/wgpu#4818): Upstream wgpu allows `wgpu::BufferViewMut` to be Send.
 unsafe impl<T> Send for CpuWriteGpuReadBuffer<T> where T: bytemuck::Pod + Send + Sync {}
 
 #[allow(unsafe_code)]
-// SAFETY: TODO(https://github.com/gfx-rs/wgpu/pull/4818): Upstream wgpu allows `wgpu::BufferViewMut` to be Sync.
+// SAFETY: TODO(gfx-rs/wgpu#4818): Upstream wgpu allows `wgpu::BufferViewMut` to be Sync.
 unsafe impl<T> Sync for CpuWriteGpuReadBuffer<T> where T: bytemuck::Pod + Send + Sync {}
 
 impl<T> CpuWriteGpuReadBuffer<T>