term: clamp zero window size from Docker PTY

[tomholford]

Summary

• When Docker allocates a PTY (tty: true + stdin_open: true) but nobody is attached, TIOCGWINSZ succeeds but reports 0x0
• This causes unsigned underflow in row_l - 1 calculations throughout term.c and a decrement-underflow crash in drum when it receives %blew [0 0]
• Clamp to 80x24 defaults when ioctl returns zero dimensions (matching existing fallback when ioctl fails)
• Fix initial default row_l from 0 to 24

Resolves #159.

Test plan

• Build with zig build
• Cross-compile with zig build -Dtarget=aarch64-linux-musl -Doptimize=ReleaseFast
• Run in Docker with tty: true + stdin_open: true and no docker attach — ship boots without crashing
• Verify docker attach / detach still works after boot

Reproduction steps

# Cross-compile for Linux
zig build -Dtarget=aarch64-linux-musl -Doptimize=ReleaseFast

# Build Docker image
cp zig-out/aarch64-linux-musl/urbit docker/
docker build -t vere-test docker/

# Create volume with comet file
docker volume create test-comet
docker run --rm -v test-comet:/urbit alpine touch /urbit/test-ship.comet

# Run with tty + stdin_open, no attach (previously crashed)
docker run -d --name test-vere -t -i -v test-comet:/urbit vere-test

# Verify ship boots
docker logs -f test-vere

# Cleanup
docker stop test-vere && docker rm test-vere
docker volume rm test-comet

[Fang-]

Regardless of where the dimensions come from, zeroes in %blew are bogus, so clamping like this is good. Arguably 0 == col_l instead of !col_l would be semantically clearer, but it's C, so readers know.

[pkova]

What version of vere are you running when you blow up with your reproduction commands? We've been unable to reproduce on linux-x86_64 and linux-aarch64.