feat: new method include_regularization_terms() for l1 and l2

[mwong009]

Checklist

• There is an informative high-level description of the changes.
• The description and/or commit message(s) references the relevant GitHub issue(s).
• pre-commit is installed and set up.
• The commit messages follow these guidelines.
• The commits correspond to relevant logical changes, and there are no commits that fix changes introduced by other commits in the same branch/BR.
• There are tests covering the changes introduced in the PR.

Features

• new: method include_regularization_terms()
• new: func compute to run the model on a manual set of params
• change: calculating distributed Beta now uses percentile instead of mean
• new: model selection type (loglikelihood or validation error)

Fix

• fix l2 regularizer formula
• set init_types as a module level list and add tests
• add minmax clipping on neural net layer outputs
• temporary function for negative relu
• expression fix for random draw variables
• allow_input_downcasting always True

Documentation

• update tutorial documentation
• add docstrings to logger
• tidy up docstrings and methods
• style: minor type hints addition

Maintenance

• refactor config.add into .add() and .update() methods
• add tests for relu, logit, and exp_mov_average
• add tests for rmse, mae, loglikelihood
• add tests for errors
• add tests for derivatives
• lock numpy to version 1.25.0
• add dill package for pickling
• relax condition for stdout during training

[codecov]

Codecov Report

Merging #86 (dc0b200) into master (e82a0fa) will increase coverage by 22.65%.
Report is 23 commits behind head on master.
The diff coverage is 57.33%.

Additional details and impacted files

@@             Coverage Diff             @@
##           master      #86       +/-   ##
===========================================
+ Coverage   31.27%   53.92%   +22.65%     
===========================================
  Files          15       16        +1     
  Lines        1573     1591       +18     
===========================================
+ Hits          492      858      +366     
+ Misses       1081      733      -348     

Files Changed	Coverage Δ
pycmtensor/statistics.py	22.64% <35.00%> (+22.64%)	⬆️
pycmtensor/scheduler.py	39.13% <40.00%> (+0.14%)	⬆️
pycmtensor/defaultconfig.py	61.90% <46.15%> (-7.54%)	⬇️
pycmtensor/models/basic.py	30.45% <47.72%> (+30.45%)	⬆️
pycmtensor/expressions.py	67.77% <52.94%> (+1.23%)	⬆️
pycmtensor/functions.py	92.68% <75.00%> (+64.58%)	⬆️
pycmtensor/results.py	37.11% <83.33%> (+37.11%)	⬆️
pycmtensor/models/layers.py	92.39% <86.66%> (+92.39%)	⬆️
pycmtensor/regularizers.py	93.75% <93.75%> (ø)
pycmtensor/logger.py	91.42% <100.00%> (ø)
... and 2 more

... and 1 file with indirect coverage changes

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 094de1e...dc0b200. Read the comment docs.