fix(KIM): use I10 in FP G3 current prefactor

[marjohma]

Closes #270

Summary

• load the I10 cell-center susceptibility moment in FP_G3_j_phi
• add a manufactured regression with deliberately distinct I10 and I01 values
• keep the change isolated to KIM source and tests; no documentation files are included

Derivation check

The azimuthal-current row carries one power of parallel velocity, so its G3 prefactor uses I^(1,0), not I^(0,1).

Verification

• test_fp_kernel_prefactors
• test_integration_methods
• test_flr2_fourier_kernel
• test_kim_solver_em

All four pass locally. Independent code review reported no findings.