Transpose the preSN flux arrays in , to match the (Ntime,Nenergy) shape, same as in CCSN

[Sheshuk]

A small fix to make flux/initial_spectra, produced by preSN models have the same shape as in case of CCSN models: provided time and energy arrays, the shape of the of the result spectra should be [len(time), len(energy)].

We don't have the checks for that resulting shapes in our tests. We'll need to revise the test suite (not in this PR)

[JostMigenda]

LGTM!

(Just out of curiosity, is there any reason to prefer (Ntime, Nenergy) over (Nenergy, Ntime)? I assume it doesn’t really matter and we can just pick one convention as long as we’re consistent, right?)

[Sheshuk]

At the moment it doesn't make any difference - because in all the places like generate_time_series we use python loop over the times, generating time points one by one:
https://github.com/SNEWS2/snewpy/blob/9096246377102e9d0e850c11396fa1e62779d642/python/snewpy/snowglobes.py#L110C2-L122

But it will make difference when we'll start using the vectorized computation in the future, so I decided to make the shapes uniform the same way as most of the CCSN models