SPU Giga block bounds check

[cipherxof]

The emulator was crashing when accessing bb.target[0] of an empty vector. This code previously used std::basic_string which probably just returned some garbage value or zero without throwing an exception.

Addresses: #16715 (and probably #16654)

[elad335]

Please upload a log with this build.
I have a suspicion as to why it happens.
The bug was there, before #16261, was just hidden because string allows to access the element at size().

[elad335]

Test with #16723 too

[cipherxof]

Test with #16723 too

This PR: RPCS3.log
PR 16723: RPCS3.log (still crashes)

[digant73]

if you are able to compile RPCS3 in DEBUG mode (I use VS 2022 on Win 11), you should get an out of bound error message when booting a game during the PPU compilation. Probably there are other issues not yet fixed caused by PR #16261 or the following PRs. In the past this issue was not present

[cipherxof]

if you are able to compile RPCS3 in DEBUG mode (I use VS 2022 on Win 11), you should get an out of bound error message when booting a game during the PPU compilation. Probably there are other issues not yet fixed caused by PR #16261 or the following PRs. In the past this issue was not present

I do not get any errors while compiling PPU modules. It probably only triggers under certain circumstances like with giga spu. I can look for potential areas later, although this PR is likely not fixing the root cause as the previous code with basic_string was masking some underlying issue. Still, it probably makes sense to have some bounds checking.

[digant73]

if you are able to compile RPCS3 in DEBUG mode (I use VS 2022 on Win 11), you should get an out of bound error message when booting a game during the PPU compilation. Probably there are other issues not yet fixed caused by PR #16261 or the following PRs. In the past this issue was not present

I do not get any errors while compiling PPU modules. It probably only triggers under certain circumstances like with giga spu. I can look for potential areas later, although this PR is likely not fixing the root cause as the previous code with basic_string was masking some underlying issue. Still, it probably makes sense to have some bounds checking.

I tested on Resistance 3 and got the error. SPU (on CPU tab) set to Mega. As I said the error is reported with RPCS3 compiled in DEBUG mode. In RELEASE mode there is no popup error displayed.