Floating-point leftovers in BASOP main

Basic info

• BASOP 97a22a9d

Bug description

Parts of the IVAS libraries code still use floating-point leftovers or are not fully converted.

1. Some of them are unused and can be simply removed. E.g.

• fft_rel()
• deg2rad()
• number of constant macros

2. Some of them are employed and need more work. E.g.

• ivas_reverb_prepare_cldfb_params()
• isar_lc3plus_*.* files
• pfNoiseBuffer in LCLD split-rendering codec
• number of constant macros

How and when to deal with this issue, such that we can test and deliver a framework that is 100% in BASOP?