Change backtrace prefix of thread ID to 1-based

[kpamnany]

PR Description

Currently the output is 0-based, so we see:

...
thread (0) ijl_apply_generic at /nix/store/6lxp69px3vnal0csypfvm3lv2dwlllwv-julia-1.9.2/lib/julia/libjulia-internal.1.9.dylib (unknown line)
thread (0) do_call at /nix/store/6lxp69px3vnal0csypfvm3lv2dwlllwv-julia-1.9.2/lib/julia/libjulia-internal.1.9.dylib (unknown line)
...

In Julia, thread ID is 1-based, so this output is wrong. This PR fixes that.

This has a messy history. The following commits should be merged with this one when we next upgrade:

• 0457f8995eb49158adcab9b12d21d9d31e9e05d4
• e015382634953445155c9fa4f51aba05235e8313
• 8f409da5c564d8fd9e9a2d0fe19235a710fedd09

Checklist

Requirements for merging:

• I have opened an issue or PR upstream on JuliaLang/julia: Stack traces can be interleaved when running multithreaded JuliaLang/julia#51056
• I have removed the port-to-* labels that don't apply.
• I have opened a PR on raicode to test these changes: https://github.com/RelationalAI/raicode/pull/15745

[nickrobinson251]

This is a fix to a RAI-specific patch, right? can you link the PRs that this is building on (to help make the porting to future versions easier)? is it e015382 and #33?

[nickrobinson251]

btw, why haven't we fixed this upstream? i can't tell from JuliaLang#51056 if there's agreement on a solution or not? (and if there is, whether we just need to find the time to go do it?)

[kpamnany]

We perhaps should do this upstream, but as you observe, there is some disagreement in the upstream issue. I'm not sure I like this solution very much -- it does make backtraces harder to read. I'll open a JIRA ticket to upstream this.

[kpamnany]

https://relationalai.atlassian.net/browse/RAI-17207