fix(bootstrap): raise tokio worker stack to 16 MiB

[EnRaiha]

Problem

Tokio worker threads use a 2 MiB default stack. The DDL → index planning path nests deep enough to overflow it, causing a silent SIGSEGV with no error message. Affected operations:

• `CREATE VECTOR INDEX`
• `CREATE SORTED INDEX`

The process dies mid-planning with `tokio-rt-worker has overflowed its stack` (or silently on the server wire path).

Reproduction

git clone https://github.com/NodeDB-Lab/nodedb.git
cd nodedb
RUST_MIN_STACK=\"\" cargo test -p nodedb --test inproc sorted_index_rows -- --test-threads 1
# → thread 'tokio-rt-worker' has overflowed its stack
# → SIGABRT, EXIT:101

Tested on x86_64 (Intel Core i5-8400, 32 GiB RAM, Debian 13 trixie / Proxmox VE). aarch64 CI runners reproduce with less nesting depth.

Fix

Raise the tokio worker stack from 2 MiB to 16 MiB in `nodedb/src/main.rs`. This matches the main-thread stack the rest of the process already relies on. Tokio reserves the space virtually and commits pages on demand, so idle workers cost nothing.

let runtime = tokio::runtime::Builder::new_multi_thread()
    .thread_stack_size(16 * 1024 * 1024)  // 16 MiB — fixes #312
    .enable_all()
    .build()?;

Verification

Config	Result
Pre-fix + no `RUST_MIN_STACK`	CRASH — stack overflow
Post-fix + no `RUST_MIN_STACK`	PASS (server binary)
Post-fix + `RUST_MIN_STACK=32MiB`	PASS (16/16 inproc tests)

Note: `#[tokio::test]` creates an independent runtime per test, so the test suite still requires `RUST_MIN_STACK=33554432` (already set in CI). The fix covers the server binary, which is what runs in production.

Files changed

• `nodedb/src/main.rs` — add `thread_stack_size(16 MiB)` to the tokio runtime builder

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.