[Compiler-rt] Tracking Issue: Use LLVM Libc math routines in builtins

This issue tracks the progress of the using of libc math routines in compiler-rt builtins.

RFC: https://discourse.llvm.org/t/rfc-explore-using-llvm-libc-math-routines-for-compiler-rt-floating-point-builtins/89670

- **Quad-Precision (128-bit / `tf`) Arithmetic**
- [x] `__addtf3` #197950
- [x] `__subtf3` #207092
- [x] `__multf3` #207092
- [x] `__divtf3` #207092

- **Quad-Precision Comparisons**
- [ ] `__letf2` #212653
- [ ] `__getf2` #212653
- [ ] `__unordtf2` #212653
- [ ] `__eqtf2` #212653
- [ ] `__netf2` #212653
- [ ] `__lttf2` #212653
- [ ] `__gttf2` #212653

- **Double-Precision Comparisons**
- [ ] `__ledf2` #212652
- [ ] `__gedf2` #212652
- [ ] `__unorddf2` #212652
- [ ] `__eqdf2` #212652
- [ ] `__nedf2` #212652
- [ ] `__ltdf2` #212652
- [ ] `__gtdf2` #212652

- **Single-Precision Comparisons**
- [ ] `__lesf2` #212651
- [ ] `__gesf2` #212651
- [ ] `__unordsf2` #212651
- [ ] `__eqsf2` #212651
- [ ] `__nesf2` #212651
- [ ] `__ltsf2` #212651
- [ ] `__gtsf2` #212651

- **Int -> Quad-Precision Conversions**
- [ ] `__floatsitf` #215725
- [ ] `__floatditf` #215725
- [ ] `__floattitf` #215725
- [ ] `__floatunsitf` #215725
- [ ] `__floatunditf` #215725
- [ ] `__floatuntitf` #215725

- **Quad-Precision -> Int Conversions**
- [ ] `__fixtfsi` #215726
- [ ] `__fixtfdi` #215726
- [ ] `__fixtfti` #215726
- [ ] `__fixunstfsi` #215726
- [ ] `__fixunstfdi` #215726
- [ ] `__fixunstfti` #215726

- **Float <-> Float Extend/Truncate Conversions**
- [x] `__extendsftf2` #209984
- [x] `__extenddftf2` #209984
- [x] `__extendxftf2` #209984
- [x] `__trunctfdf2` #209984
- [x] `__trunctfsf2` #209984
- [x] `__trunctfxf2` #209984

- **Half-Precision (`_Float16`) Conversions**
- [ ] `__extendhfsf2` #211882
- [ ] `__extendhfdf2` #211882
- [ ] `__extendhftf2` #211882
- [ ] `__truncsfhf2` #211882
- [ ] `__truncdfhf2` #211882
- [ ] `__trunctfhf2` #211882

- **BFloat16 Conversions**
- [ ] `__truncsfbf2` #211881
- [ ] `__truncdfbf2` #211881
- [ ] `__trunctfbf2` #211881
- [ ] `__extendbfsf2` #211881

- **Double-Precision Arithmetic (Soft-Float Targets)**
- [x] `__adddf3` #207092
- [x] `__subdf3` #207092
- [x] `__muldf3` #207092
- [x] `__divdf3` #207092
- [x] `__negdf2` #207092

- **Single-Precision Arithmetic (Soft-Float Targets)**
- [x] `__addsf3` #207092
- [x] `__subsf3` #207092
- [x] `__mulsf3` #207092
- [x] `__divsf3` #207092
- [x] `__negsf2` #207092

- **x87 80-bit Extended Precision**
- [ ] `__floatdixf` #215729
- [ ] `__floattixf` #215729
- [ ] `__floatundixf` #215729
- [ ] `__floatuntixf` #215729
- [ ] `__fixunsxfsi` #215729
- [ ] `__fixunsxfdi` #215729
- [ ] `__fixunsxfti` #215729
- [ ] `__fixxfdi` #215729
- [ ] `__fixxfti` #215729
- [ ] `__extendhfxf2` #215729
- [ ] `__truncxfhf2` #215729
- [ ] `__truncxfbf2` #215729

- **Int -> Double / Int -> Single Conversions**
- [x] `__floatsidf` #209900
- [x] `__floatdidf` #209900
- [x] `__floattidf` #209900
- [x] `__floatsisf` #209900
- [x] `__floatdisf` #209900
- [x] `__floattisf` #209900
- [x] `__floatunsidf` #209900
- [x] `__floatundidf` #209900
- [x] `__floatuntidf` #209900
- [x] `__floatunsisf` #209900
- [x] `__floatundisf` #209900
- [x] `__floatuntisf` #209900

- **Double/Single -> Int Conversions**
- [x] `__fixdfsi` #207543
- [x] `__fixdfdi` #207543
- [x] `__fixdfti` #207543
- [x] `__fixsfsi` #207543
- [x] `__fixsfdi` #207543
- [x] `__fixsfti` #207543
- [x] `__fixunsdfsi` #207543
- [x] `__fixunsdfdi` #207543
- [x] `__fixunsdfti` #207543
- [x] `__fixunssfsi` #207543
- [x] `__fixunssfdi` #207543
- [x] `__fixunssfti` #207543

- **Double <-> Single Extend/Truncate**
- [x] `__extendsfdf2` #209984
- [x] `__truncdfsf2` #209984

CC: @michaelrj-google @bassiounix @lntue