build: remove redundant inline specifiers from implicitly inline functions

[bobtista]

• Used clang-tidy

[Caball009]

Are we sure that VC6 treats inline the same way as modern compilers do? I think we've seen in the past that different function inlining can cause mismatches for seemingly obscure reasons.

I'd recommend running this change against a large number of replays.

[bobtista]

VC6 treats inline the same way as modern compilers do

Section 7.1.2: "A function defined within a class definition is an inline function."
C++98 Standard (ISO/IEC 14882:1998)

This means member functions defined inside class bodies are implicitly inline in C++98, regardless of the inline keyword. VC6 claims C++98 support, so it should follow this. How many replays does our CI check with?

[xezon]

If they were not implict inline then it would compile error, because of duplicate symbol definitions.