Add index_base as optional argument to export_data

[jeremy-myers]

export_data currently writes out sparse index sets in 1-based indexing, but import_data allows any arbitrary base. It's not obvious that we enforce one base on write but allow any on read.

for i in range(A.nnz):
    # 0-based indexing in package, 1-based indexing in file
    subs = A.subs[i, :] + 1 # <- Here we assume 1-based indexing