Feature&Refactor: support SCF restarts from given density matrices and several minor fixes

[kirk0830]

What's changed

1. disable the warnings in lead_property.py
2. connect the ele_T (electronic temperature) user settings with smearing
3. support the externally given pcond as the initial guess of the Poisson-NEGF SCF
4. extend the upper limit on the number of regions by 1

[Copilot]

Pull request overview

This PR updates the NEGF workflow to better support SCF restarts from an externally prepared Poisson interface, while also aligning electronic-temperature settings with Fermi-level smearing and applying a few minor configuration/doc fixes.

Changes:

• Wire ele_T (electronic temperature) into Fermi-level smearing temperature during lead Fermi-level calculation.
• Add optional pcond: Interface3D input to NEGF.compute() to allow starting Poisson-NEGF SCF from an externally provided Poisson state (and return it for reuse).
• Extend Poisson dielectric-region input support to allow one additional dielectric region; silence some HDF5 “already exists” warnings.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 4 comments.

File	Description
dpnegf/utils/argcheck.py	Extends accepted dielectric region keys (supports one more region).
dpnegf/runner/NEGF.py	Uses ele_T for smearing temperature; adds compute(pcond=...) restart path and return value.
dpnegf/negf/poisson_init.py	Minor docstring whitespace/formatting adjustments.
dpnegf/negf/lead_property.py	Comments out warnings during HDF5 write/merge for self-energy caching.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

write_to_hdf5 no longer guards against an existing dataset name. With h5py.File(..., "a"), reruns/resumes can hit the same (E,k) and grp.create_dataset(...) will raise (dataset already exists), aborting the self-energy workflow. If the goal is to silence warnings, keep the existence check but skip (or explicitly overwrite via del grp[dset_name] / use require_dataset).

Suggested change

	# if dset_name in grp:
	# log.warning(f"Dataset {dset_name} already exists in group {group_name}. Skipping it.")
	if dset_name in grp:
	log.warning(f"Dataset {dset_name} already exists in group {group_name}. Skipping it.")
	return

[AsymmetryChou]

All checks.