Raise error if broadcasting with MottonenStatePreparation and BasisStatePreparation compute_decomposition

[mudit2812]

Context:
MottonenStatePreparation and BasisStatePreparation fails right now when decomposing a broadcasted state vector because different state vectors have different decompositions.

Description of the Change:

• Raise an error in MottonenStatePreparation.compute_decomposition and BasisStatePreparation.compute_decomposition if the batch_size if not None and suggest users to use broadcast_expand.

Benefits:
No failures or wrong results when decomposing Mottonen or BasisState

Possible Drawbacks:

Related GitHub Issues:
#4753

[mudit2812]

[sc-48950]

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (deaf387) 99.64% compared to head (d404845) 99.63%.
Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #4767      +/-   ##
==========================================
- Coverage   99.64%   99.63%   -0.02%     
==========================================
  Files         381      381              
  Lines       34336    34121     -215     
==========================================
- Hits        34214    33996     -218     
- Misses        122      125       +3     

Files	Coverage Δ
pennylane/templates/state_preparations/basis.py	100.00% <100.00%> (ø)
pennylane/templates/state_preparations/mottonen.py	100.00% <100.00%> (ø)
pennylane/transforms/transpile.py	100.00% <100.00%> (ø)

... and 43 files with indirect coverage changes

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

[albi3ro]

broadcast_expand does not fix the problem.

The problem is that op.batch_size will still be None, so broadcast_expand will not register that the template has a batched.

broadcast_expand does work if we set ndim_params = (1,) for MottonenStatePrep.

[albi3ro]

👍