perf: improve GlobPath performance

[letFunny]

Avoid expensive calls to GlobPath by separating the string into segments
and using cheaper comparisons were possible.

• Have you signed the CLA?

---

Depends on #280, #281 and #279.

[letFunny]

The CI that comments in not working at the moment. This is the result of the performance benchmark:

Benchmark 1: BASE
  Time (mean ± σ):     14.034 s ±  0.232 s    [User: 14.542 s, System: 0.081 s]
  Range (min … max):   13.795 s … 14.490 s    10 runs
 
Benchmark 2: HEAD
  Time (mean ± σ):      1.843 s ±  0.016 s    [User: 1.997 s, System: 0.033 s]
  Range (min … max):    1.828 s …  1.877 s    10 runs
 
Summary
  'HEAD' ran
    7.61 ± 0.14 times faster than 'BASE'

Which means in total that all the improvements make the algorithm 7.61 times faster.