static float f = 0.f / 0.f; cannot be compiled

This program will not compile.

```c
int main(void){
if(1){
static float f = 0.f / 0.f;/* cannot compile this constant l-value expression yet */
}
return 0;
}
```

I believe that C99 introduced this feature along with IEEE-754 floating-point.
It is one way to create a NAN.

No source file or test is named. Start by compiling the minimal C reproducer and checking the diagnostic against C99 constant-expression requirements, then trace handling of static initializers and floating-point constant expressions. Done means the example is accepted as intended and has regression coverage.

Written by the indexing model from the issue text.