Change unit test response file to dense

[israelmcmc]

We have moved from a sparse to a dense response. It was not that sparse, and it ended up being slower than a dense matrix. The unit test now reflects what we actually use.

I created the dense response with:

from cosipy.response import FullDetectorResponse
from cosipy import test_data
from histpy import Histogram

dr = FullDetectorResponse.open(test_data.path/'test_full_detector_response.h5')

dr_new = Histogram.concatenate(dr.axes['NuLambda'], list(dr))
dr_new = dr_new.project(['Ei', 'NuLambda', 'Em', 'Phi', 'PsiChi'])
dr_new = dr_new.todense()

FullDetectorResponse._write_h5(dr_new, test_data.path/'test_full_detector_response_dense.h5')

In order to make things easier I moved the code that writes an HDF5 response file from a Histogram from _open_rsp to its own function, _write_h5. I added a unit test for this specific function.

[codecov]

Codecov Report

Attention: Patch coverage is 35.48387% with 20 lines in your changes missing coverage. Please review.

Project coverage is 79.58%. Comparing base (aea580b) to head (382a503).

Files with missing lines	Patch %	Lines
cosipy/response/FullDetectorResponse.py	35.48%	20 Missing ⚠️

❌ Your patch check has failed because the patch coverage (35.48%) is below the target coverage (100.00%). You can increase the patch coverage or adjust the target coverage.

Files with missing lines	Coverage Δ
cosipy/response/FullDetectorResponse.py	45.99% <35.48%> (+6.70%)	⬆️

[israelmcmc]

I need to create a fake .rsp to fix the test coverage. @augustus-thomas I think at some point you were working on this, did you get to create one? Or maybe it was for the PSR, I don't remember exactly.

[hiyoneda]

maybe, should it be tmp_path.path / 'tmp_rsp.h5'? Currently, the file will be saved in the test data directory.

[israelmcmc]

Thanks for the catch. Indeed it was the wrong path. I fixed it in my latest commit.