Shepp logan phantom

[georgeyiasemis]

• Add Shepp Logan Datasets
• Add sensitivity map simulator
• Refactor direct train, direct predict and python3 projects/predict_val.py to not necessarily require path to data as some datasets don't require it - build_dataset_from_input relies on **kwargs now
• Refactored configs and docs to comply with the above

[codecov]

Codecov Report

Merging #202 (11d203d) into main (bceae48) will increase coverage by 0.10%.
The diff coverage is 84.41%.

@@            Coverage Diff             @@
##             main     #202      +/-   ##
==========================================
+ Coverage   80.69%   80.79%   +0.10%     
==========================================
  Files          71       72       +1     
  Lines        5096     5317     +221     
==========================================
+ Hits         4112     4296     +184     
- Misses        984     1021      +37     

Impacted Files	Coverage Δ
direct/cli/predict.py	0.00% <0.00%> (ø)
direct/cli/train.py	0.00% <0.00%> (ø)
direct/common/subsample.py	70.85% <ø> (ø)
direct/inference.py	0.00% <0.00%> (ø)
direct/predict.py	0.00% <ø> (ø)
direct/cli/utils.py	60.00% <50.00%> (+2.10%)	⬆️
direct/train.py	56.93% <53.65%> (-9.43%)	⬇️
direct/data/h5_data.py	64.82% <76.92%> (-0.14%)	⬇️
direct/data/datasets.py	88.45% <96.73%> (+5.82%)	⬆️
direct/data/datasets_config.py	100.00% <100.00%> (ø)
... and 6 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update bceae48...11d203d. Read the comment docs.

[jonasteuwen]

Looks good to me, but please add references where possible, e.g. in simulating a sensitivity map.