fix(eval): iterative evaluation improvements; SWE-Bench multimodal fixes

[xingyaoww]

• This change is worth documenting at https://docs.all-hands.dev/
• Include this change in the Release Notes. If checked, you must provide an end-user friendly description for your change below

End-user friendly description of the problem this fixes or functionality that this introduces.

---

Give a summary of what the PR does, explaining any non-trivial design decisions.

• Improve iterative evaluation to retry on empty patch
• Make some of eval logging as debug level
• Fix json.loads issue for SWE-Bench Multimodal instances
• Handle SWE-Bench init script in different location (for multimodal)

---

Link of any specific issues this addresses.

---

To run this PR locally, use the following command:

docker run -it --rm   -p 3000:3000   -v /var/run/docker.sock:/var/run/docker.sock   --add-host host.docker.internal:host-gateway   -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:8160f2b-nikolaik   --name openhands-app-8160f2b   docker.all-hands.dev/all-hands-ai/openhands:8160f2b

[juanmichelini]

LGTM!

[xingyaoww]

Did a quick validation evaluation run, this give us 30/50. Good to merge!