Fix #4588 (crash in Control::_set_anchor)

[bojidar-bg]

@vnen @her001 ~ You might want to test this PR 😄

[vnen]

Still segfaults in release mode, though it's not really related to your code. Gives me the same problem without this patch even after I revert 0e8a8d2 and bac875b. I'll try to bisect this.

(gdb) where
#0  0x0000000000594937 in DVector<String>::resize(int) ()
#1  0x00000000011fdd8c in ResourceInteractiveLoaderBinary::parse_variant(Variant&) ()
#2  0x00000000011fc669 in ResourceInteractiveLoaderBinary::parse_variant(Variant&) ()
#3  0x00000000011ff94b in ResourceInteractiveLoaderBinary::poll() ()
#4  0x000000000122692a in ResourceFormatLoader::load(String const&, String const&, Error*)
    ()
#5  0x00000000012260b9 in ResourceLoader::load(String const&, String const&, bool, Error*)
    ()
#6  0x0000000000429ff3 in Main::start() ()
#7  0x000000000040d0ed in main ()

[vnen]

Still getting segfault at 37af8b4. How come nobody ever got this bug? Or is it just me?

[akien-mga]

@bojidar-bg Should we merge this patch as a general code quality improvement even if it doesn't fix #4588? (if so maybe you could amend the commit log, or alternatively I could do a "squash and merge" and rewrite the log).

[vnen]

@akien-mga technically, this does fix #4588. But it seems there's another bug with something else that gives segfaults in the GUI demos in release builds. Such bug seems not to be related to this patch or with this feature in general. I will open another issue about it.

[bojidar-bg]

@akien-mga ~ well, feel free to rewrite and merge, though it probably does fix the issue (since master without that feature still segfaluts on release).

[akien-mga]

Thanks, let's merge then.

[vnen]

Opened #4623 about the problem.