Reduce unnecessary COW on Vector by make writing explicit

This commit makes operator[] on Vector const and adds a write proxy to it.  From
now on writes to Vectors need to happen through the .write proxy. So for
instance:

Vector<int> vec;
vec.push_back(10);
std::cout << vec[0] << std::endl;
vec.write[0] = 20;

Failing to use the .write proxy will cause a compilation error.

In addition COWable datatypes can now embed a CowData pointer to their data.
This means that String, CharString, and VMap no longer use or derive from
Vector.

_ALWAYS_INLINE_ and _FORCE_INLINE_ are now equivalent for debug and non-debug
builds. This is a lot faster for Vector in the editor and while running tests.
The reason why this difference used to exist is because force-inlined methods
used to give a bad debugging experience. After extensive testing with modern
compilers this is no longer the case.

1 files changed, 3 insertions, 3 deletions

diff --git a/platform/iphone/export/export.cpp b/platform/iphone/export/export.cpp
index 5480d30e7a..e59c81a148 100644
--- a/platform/iphone/export/export.cpp
+++ b/platform/iphone/export/export.cpp
@@ -276,7 +276,7 @@ void EditorExportPlatformIOS::_fix_config_file(const Ref<EditorExportPreset> &p_
 	CharString cs = strnew.utf8();
 	pfile.resize(cs.size() - 1);
 	for (int i = 0; i < cs.size() - 1; i++) {
-		pfile[i] = cs[i];
+		pfile.write[i] = cs[i];
 	}
 }
 
@@ -489,7 +489,7 @@ private:
 		ret.resize(sizeof(num) * 2);
 		for (int i = 0; i < sizeof(num) * 2; ++i) {
 			uint8_t four_bits = (num >> (sizeof(num) * 8 - (i + 1) * 4)) & 0xF;
-			ret[i] = _hex_char(four_bits);
+			ret.write[i] = _hex_char(four_bits);
 		}
 		return String::utf8(ret.ptr(), ret.size());
 	}
@@ -587,7 +587,7 @@ void EditorExportPlatformIOS::_add_assets_to_project(Vector<uint8_t> &p_project_
 	CharString cs = str.utf8();
 	p_project_data.resize(cs.size() - 1);
 	for (int i = 0; i < cs.size() - 1; i++) {
-		p_project_data[i] = cs[i];
+		p_project_data.write[i] = cs[i];
 	}
 }