Feat: add ZBL weighted DP model

[anyangml]

This PR is to implement general linear combination of several DPAtomicModels with user defined weights, as well as a special case ZBLModel with weights calculated based on this paper: Appl. Phys. Lett. 114, 244101 (2019); doi: 10.1063/1.5098061

[njzjz]

I haven't carefully checked - but could you first add a test for JIT?

deepmd-kit/source/tests/pt/model/test_jit.py

Line 34 in 79f98ca

model = torch.jit.script(inference.Tester("./model.pt", numb_test=1).model)

Have jit passed?

[anyangml]

@njzjz Jit passed, check the UTs in this file source/tests/pt/model/test_linear_atomic_model.py, I believe all the data are kept when you jit a deserialized model. Please double check this code block

deepmd-kit/deepmd/pt/model/model/pairtab_atomic_model.py

Lines 50 to 84 in ccf63ce

	def __init__(
	self, tab_file: str, rcut: float, sel: Union[int, List[int]], **kwargs
	):
	super().__init__()
	self.tab_file = tab_file
	self.rcut = rcut
	self.tab = self._set_pairtab(tab_file, rcut)
	# handle deserialization with no input file
	if self.tab_file is not None:
	(
	tab_info,
	tab_data,
	) = self.tab.get() # this returns -> Tuple[np.array, np.array]
	self.tab_info = torch.from_numpy(tab_info)
	self.tab_data = torch.from_numpy(tab_data)
	else:
	self.tab_info = None
	self.tab_data = None
	# self.model_type = "ener"
	# self.model_version = MODEL_VERSION ## this shoud be in the parent class
	if isinstance(sel, int):
	self.sel = sel
	elif isinstance(sel, list):
	self.sel = sum(sel)
	else:
	raise TypeError("sel must be int or list[int]")
	@torch.jit.ignore
	def _set_pairtab(self, tab_file: str, rcut: float) -> PairTab:
	return PairTab(tab_file, rcut)