Add conditional LogZ calculation

[josephdviviano]

When using a conditional gflownet, we need the LogZ estimate (in algorithms such as TB) to be conditional, i.e., predicted from the state.

This means the logz parameter must not be constrained to be a single parameter.

Also, for conditional generation, sampling should accept an instance and/or iterator optionally, to condition on.

[josephdviviano]

Addressed in https://github.com/GFNOrg/torchgfn/pull/188