fix bug with CNOT/PauliZ, TF and large batched states

[timmysilv]

Context:
First reported on the forum. qml.CNOT uses apply_operation_tensordot when that state has >=9 dims and the interface is tensorflow, but the is_state_batched param was missing (so it always assumed it wasn't batched). While fixing this, I noticed that it was also missing with PauliZ.

Description of the Change:
Pass the correct is_state_batched value to apply_operation_tensordot when called from the CNOT and PauliZ apply functions.

Benefits:
No more unexpected errors. I think sometimes they would not raise error, but give unexpected results. That shouldn't happen either now.

Possible Drawbacks:
N/A

Fixes #4892

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (34a561a) 99.65% compared to head (b2e161e) 99.64%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #4889      +/-   ##
==========================================
- Coverage   99.65%   99.64%   -0.02%     
==========================================
  Files         388      388              
  Lines       34964    34700     -264     
==========================================
- Hits        34844    34577     -267     
- Misses        120      123       +3     

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