Add Context-Sensitive IR PGO (CSIR PGO)

Clang has support for an additional PGO mode - Context-Sensitive PGO. This is completely the same way as now used by Rustc but the main difference is when instrumentation is done. In usual (IR PGO) instrumentation is done before the inlining phase, with CSIR PGO - after the inlining phase. This could be important since the inlining decisions can introduce some noise into the profiling information.

In https://github.com/llvm/llvm-project/issues/56274#issuecomment-1407117774 comment, there is a small insight into the actual effects on the performance. According to this, we can treat CSIR PGO as a light replacement for LLVM BOLT optimization. For some people, it could be important since LLVM BOLT does not work on all platforms.

Right now it's not clear, could we implement CSIR PGO by just passing the right LLVM flag, or some frontend changes are required too? I guess some CSIR PGO implementation in Clang is required here. In any way, CSIR PGO information should be added to the Rustc PGO documentation.

CSIR PGO also was mentioned in the initial PGO issue for Rustc: https://github.com/rust-lang/rust/issues/59913#issue-432504990

Start with the Rustc PGO documentation and Clang's UsersManual CS profile-generation section, then review the linked Rust and LLVM issue discussions. Determine whether CSIR PGO needs Rustc frontend changes or only an LLVM flag; done means the mode works in Rustc and its documentation explains how to use it.