Free memory after initializing random_phi during first training iteration

[roi-meir]

Summary

During the first training iteration (when reinit is true), scHPF allocates a temporary random_phi Dirichlet matrix to initialize Xphi_data. That array is not deleted afterward, so it can linger until the next GC cycle and inflate peak memory use—especially with large batches / many nonzeros.

Proposed fix

Delete random_phi immediately after it is used to construct Xphi_data:

Xphi_data = X_batch.data[:,None] * random_phi
del random_phi  # free memory