all: Enforce weak / global functions

[german77]

This is a continuation of #262. Fully eliminates all weak / global mismatches from sead. Unlike the previous PR I call it an aggressive approach since it will break some matches and duplicate code. The lost matches will be recovered once the object is used.

---

This change is 

[MonsterDruide1]

@MonsterDruide1 reviewed 8 files and all commit messages, and made 3 comments.
Reviewable status: 8 of 10 files reviewed, 3 unresolved discussions (waiting on german77).

---

modules/src/devenv/seadFontMgr.cpp at r1 (raw file):
File is empty now => delete it (and remove from CMakeLists.txt)

---

modules/src/math/seadMathCalcCommon.cpp at r1 (raw file):
same here

---

modules/src/prim/seadStringBuilder.cpp at r1 (raw file):
This seems wrong, and ... to my knowledge, is wrong.
Checking <char>::copy and <char16_t>::copy within Labo, which has SEAD_ASSERTs enabled, we can see that for both instantiations, the str must not be null message is generated in line 42 - so it must be coming from the same function implementation.

Please revert these ones (StringBuilder and MathCalcCommon) which introduce duplicate code - hopefully we'll figure out something in time, otherwise we'll have to come up with a "blacklist" to ignore specific issues, but introducing that much duplicate code is just wrong.

[german77]

@german77 made 1 comment.
Reviewable status: 8 of 10 files reviewed, 3 unresolved discussions (waiting on MonsterDruide1).

---

modules/src/prim/seadStringBuilder.cpp at r1 (raw file):

Previously, MonsterDruide1 wrote…

This seems wrong, and ... to my knowledge, is wrong.
Checking <char>::copy and <char16_t>::copy within Labo, which has SEAD_ASSERTs enabled, we can see that for both instantiations, the str must not be null message is generated in line 42 - so it must be coming from the same function implementation.

Please revert these ones (StringBuilder and MathCalcCommon) which introduce duplicate code - hopefully we'll figure out something in time, otherwise we'll have to come up with a "blacklist" to ignore specific issues, but introducing that much duplicate code is just wrong.

I kind of agree that this isn't the correct way to do it since it makes the code awful. Thanks for the assert suggestion. I will try to find another solution and if it takes too long I will fully rever these changes.

[german77]

@german77 made 3 comments.
Reviewable status: 7 of 12 files reviewed, 3 unresolved discussions (waiting on MonsterDruide1).

---

modules/src/devenv/seadFontMgr.cpp at r1 (raw file):

Previously, MonsterDruide1 wrote…

File is empty now => delete it (and remove from CMakeLists.txt)

Done.

---

modules/src/math/seadMathCalcCommon.cpp at r1 (raw file):

Previously, MonsterDruide1 wrote…

same here

Done. Sadly gcd now has a regswap

---

modules/src/prim/seadStringBuilder.cpp at r1 (raw file):

Previously, german77 (Narr the Reg) wrote…

I kind of agree that this isn't the correct way to do it since it makes the code awful. Thanks for the assert suggestion. I will try to find another solution and if it takes too long I will fully rever these changes.

Done. Now every function has an Impl_ equivalent. I made sure no unwanted symbols are generated in this conversion.