[clang] [clang][AST] Handle dependent representation of call to function with explicit object parameter in CallExpr::getBeginLoc() (PR #126868)

[Younan Zhang via cfe-commits]

zyn0217 wrote:

> > While this may not be the most optimal solution, we can likely go with it as-is to put the fire out.
> 
> Given [this comment](https://github.com/llvm/llvm-project/pull/126868#discussion_r1953948099) (i.e. that the behaviour with this patch is correct for the dependent case as well), is there anything else not optimal about this solution?

I was considering finding a better AST model, but it's more complex than the approach so this is good enough for now :P


https://github.com/llvm/llvm-project/pull/126868