Plotting methods for BassModel

[williambdean]

Add dedicated plotting methods to BassModel for visualizing adoption curves and diagnostics.

Motivation

The existing Bass notebook shows several useful plots, but users must build them manually. Adding model.plot_* methods (consistent with MMM and CLV patterns) makes analysis easier and enables MLflow integration via mlflow.log_figure.

Proposed plots

Method	Description
plot_adoption_curve	Adoption curve: observed vs. predicted adopters over time with HDI
plot_cumulative	S-curve: cumulative adoption with HDI
plot_decomposition	Breakdown into innovators vs. imitators (see #1706 for dual y-axes)
plot_peak	Posterior distribution of peak adoption time

Design notes

• Follow the pattern in pymc_marketing/clv/plotting.py (class methods that accept ax and figsize)
• Support both single-product and multi-product data (with product parameter to select)
• Use ArviZ for HDI computation where applicable
• Reuse F and f functions for theoretical curves

Related

• Dual y-axes for the "Innovators vs Imitators" plots in bass notebook #1706 - Dual y-axes for innovator vs. imitator plots
• MLflow integration for BassModel #2588 - MLflow integration depends on this

[williambdean]

Connected issues:

• Update Bass notebook to use BassModel class #2590 — Update Bass notebook depends on this for plotting methods
• MLflow integration for BassModel #2588 — MLflow integration depends on this for logged figures
• Dual y-axes for the "Innovators vs Imitators" plots in bass notebook #1706 — Dual y-axes for innovator vs. imitator plots (should be addressed here)