Expand Machine Code Analysis page in the performance guide

[GabrielMajeri]

This is a tracking issue for expanding upon the machine code analysis part of the perf guide.

To do

• [ ] how to emit assembly from Rust code (e.g. cargo-asm)
• [ ] how to emit IACA / llvm-mca markers with inline assembly
• [ ] give an example on how to intepret results, in the context of one of the example benchmarks

References

• IACA homepage
• "What is IACA and how to use it?" on StackOverflow
• TheIronBorn's usage of IACA with movmsk.c
• LLVM MCA docs

[hdevalence]

I made a crate last year that makes using IACA slightly easier: https://github.com/hdevalence/iaca-marker-macros