Perf: Add OpenMP parallel for GintInfo::init_atoms_()

[amnesi-ary]

Reminder

• Have you linked an issue with this pull request?
• Have you added adequate unit tests and/or case tests for your pull request?
• Have you noticed possible changes of behavior below or in the linked issue?
• Have you explained the changes of codes in core modules of ESolver, HSolver, ElecState, Hamilt, Operator or Psi? (ignore if not applicable)

What's changed?

• Split orbital copying from atom processing so per-atom work can run independently.
• Switch atom traversal to iterate by global atom index using iat2it and iat2ia.
• Added OpenMP parallelization with dynamic scheduling for the per-atom big-grid search.

Motivation:
Resolve the TODO in GintInfo::init_atoms_() by parallelizing the atom initialization path.

[amnesi-ary]

label: project-learning