Add Censored wrapper for Prior class

[wd60622]

Closes #1238

Related to #1296

from pymc_marketing.prior import Prior, Censored

normal = Prior("Normal", dims=("channel", "geo"))
censored_normal = Censored(normal, lower=0)

With implemented methods:

• create_variable
• create_likelihood_variable
• sample_prior
• to_graph
• to_dict
• from_dict

And support from deserialize function

This allows it to be used in MMM. For example:

from pymc_marketing.mmm import MMM

normal = Prior("Normal", sigma=Prior("HalfNormal"), dims="date")
censored_normal = Censored(normal, lower=0) 

model_config = {"likelihood": censored_normal}

mmm = MMM(
    ...,
    model_config=model_config,
)

CC: @drbenvincent

---

📚 Documentation preview 📚: https://pymc-marketing--1309.org.readthedocs.build/en/1309/

[codecov]

Codecov Report

Attention: Patch coverage is 92.95775% with 5 lines in your changes missing coverage. Please review.

Project coverage is 95.10%. Comparing base (8600ef3) to head (b6a0457).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
pymc_marketing/prior.py	92.95%	5 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1309      +/-   ##
==========================================
- Coverage   95.14%   95.10%   -0.04%     
==========================================
  Files          44       44              
  Lines        4633     4703      +70     
==========================================
+ Hits         4408     4473      +65     
- Misses        225      230       +5     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[juanitorduz]

cool! just a suggestion about the class type :)