tl;dr section suggests options that it maybe shouldn't

[AaronBallman]

Clang doesn't support -Wtrampolines and will diagnose it as an unknown warning option, so it might not make sense to include it in the tl;dr section (I would expect some people will copy/paste that section without giving it much thought).

Also, _FORTIFY_SOURCE support differs between Clang and GCC (https://developers.redhat.com/blog/2020/02/11/toward-_fortify_source-parity-between-clang-and-gcc) and those differences should be called out explicitly in the tl;dr so people don’t get a false sense of security with the option.

[thomasnyman]

Possibly partial duplicate of #277.

The feature parity issue is addressed by footnote referenced from the main table but I see that footnote seems to have gathered a couple of erroneous references, probably due to copy-pasting. Let's fix those and make sure it also is linked from the tl;dr;.

[david-a-wheeler]

We now have a table showing "if condition X, use option Y". I think it's fine to have a gcc-specific entry (or a clang-specific entry) if an option is supported by one but not the other.