[non-BE][split-non-BE][allow-regression] Resolve "Probably porting mistake in protoSignalComputation_shd_fx"

• Related issues: #2511 (closed)
• Requested reviewers: @tritthart

Reason why this change is needed

• The BASOP code of protoSignalComputation_shd() wasn't ported correctly:

  reference_power[l + ( k + 1 ) * num_freq_bands] += *( p_k[k] ) * *( p_k[k] );

• Instead of the +=, only a regular assignmed was implemented

  reference_power_fx[idx1] = Mpy_32_32( *p_k_fx[k], *p_k_fx[k] ); // Q(2*(q_cldfb + min_q_shift)-31)

Description of the change

• Use Madd_32_32() instead of Mpy_32_32()

Affected operating points

• Describe here as well as possible what operating points are affected and how
• In minimum, there should be a status for bitstream compatibility and output bit exactness
• For bitstream compatibility, the following levels are helpful for describing encoder
  1. Produced bitstream is BE compared to previous state.
  2. Produced bitstream is non-BE but it is fully backwards compatible for decoding. Decoded output may differ.
  3. Produced bitstream is non-BC. Old decoder cannot decode the produced bitstream correctly.
• For output difference, use BE or non-BE. Additionally, amount of difference can be presented.

Closes #2511 (closed)