Docking of metal ions is not straightforward. Most [open source] docking tools lack the required atomic parameters of metal ions. You'll need to provide the specific atomic parameters for metal ion docking. AutoDock Vina and AutoDock 4.0 are good options among the open source tools. You can look these previous threads with the same question:

https://www.researchgate.net/post/I_am_trying_to_dock_As3_ions_in_a_protein_called_DJ-1_Autodock_is_showing_errors_while_docking_How_to_go_about_it

https://www.researchgate.net/post/How_to_perform_protein_docking_with_metal_ions_using_autodock.

This can be done with "classic" AutoDock 4.2.6. However, the results might not be biochemically meaningful, as the most favorable binding sites will likely be dominated by electrostatic interactions. It is also important to realize that docking of a bare metal ion would not reflect the complexation state of the metal in a biological medium, where the metal would exist as a complex with various ligands found in solution, e.g., amino acids, peptides, proteins, and small anions.

This is a experience I would share; you can simply find a pdb file from a database which has a calcium ion and copy-past its atom types and coordinates into your pdf file. However, the ion may not be resides within the appropriate site within the protein. For fixing such complication you can manually change x y z coordination by changing 1.000, 2.000, and 3.000 values in the following line of CA structure in a pdb file;

HETATM 1 CA CAL A 1 1.000 2.000 3.000 1.00 0.00 CA

For docking you may use DOCK Blaster server, but bling docking dose not determine the correct binding site.