Fix first user message

[enyst]

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

• Include this change in the Release Notes. If checked, you must provide an end-user friendly description for your change below
  Fix DeepSeek error on consecutive user messages.

---

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

Users are getting an error about "consecutive" user messages with the new Deepseek. Reported and WIP on the litellm repo:

• [Bug]: [DeepSeek R1] 400 error for 2 consecutive messages with role user BerriAI/litellm#7972

{'error': {'message': 'deepseek-reasoner does not support successive user or assistant messages

This PR proposes to refactor the first user message more aligned with the other prompt enhancement we have, and fix it for us, anyway, while we're at it.

---

Link of any specific issues this addresses
Fix #6459
Fix #6443
Reported on slack

---

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:21d8c00-nikolaik   --name openhands-app-21d8c00   docker.all-hands.dev/all-hands-ai/openhands:21d8c00

[github-actions]

Hi! I started running the integration tests on your PR. You will receive a comment with the results shortly.

[github-actions]

Trigger by: Pull Request (integration-test label on PR #6471)
Commit: 274b3a9
Integration Tests Report (Haiku)
Haiku LLM Test Results:
Success rate: 100.00% (7/7)

Total cost: USD 0.11

instance_id	success	reason	cost	error_message
t01_fix_simple_typo	True		0.015	nan
t06_github_pr_browsing	True		0.03	nan
t03_jupyter_write_file	True		0.014	nan
t07_interactive_commands	True		0.016	nan
t02_add_bash_hello	True		0.015	nan
t04_git_staging	True		0.013	nan
t05_simple_browsing	True		0.009	nan

---

Integration Tests Report (DeepSeek)
DeepSeek LLM Test Results:
Success rate: 0.00% (0/7)

Total cost: USD 0.00

instance_id	success	reason	cost	error_message
t06_github_pr_browsing	False	The answer is not found in any message. Total messages: 1.	0	RuntimeError: There was an unexpected error while running the agent. Please report this error to the developers. Your session ID is default. Error type: Timeout
t01_fix_simple_typo	False	File not fixed: This is a stupid typoo.	0	RuntimeError: There was an unexpected error while running the agent. Please report this error to the developers. Your session ID is default. Error type: Timeout
		Really?
		No mor typos!
		Enjoy!
t07_interactive_commands	False	The answer is not found in any message. Total messages: 1.	0	RuntimeError: There was an unexpected error while running the agent. Please report this error to the developers. Your session ID is default. Error type: Timeout
t03_jupyter_write_file	False	Failed to cat /workspace/test.txt: cat: /workspace/test.txt: No such file or directory.	0	RuntimeError: There was an unexpected error while running the agent. Please report this error to the developers. Your session ID is default. Error type: Timeout
t02_add_bash_hello	False	Failed to cat /workspace/hello.sh: cat: /workspace/hello.sh: No such file or directory.	0	RuntimeError: There was an unexpected error while running the agent. Please report this error to the developers. Your session ID is default. Error type: Timeout
t05_simple_browsing	False	The answer is not found in any message. Total messages: 1.	0	RuntimeError: There was an unexpected error while running the agent. Please report this error to the developers. Your session ID is default. Error type: Timeout
t04_git_staging	False	Failed to check for "nothing to commit, working tree clean": On branch master	0	RuntimeError: There was an unexpected error while running the agent. Please report this error to the developers. Your session ID is default. Error type: Timeout
		No commits yet
		Changes to be committed:
		(use "git rm --cached ..." to unstage)
		new file: hello.py.

---

Integration Tests Report Delegator (Haiku)
Success rate: 50.00% (1/2)

Total cost: USD 0.00

instance_id	success	reason	cost	error_message
t02_add_bash_hello	True		0
t01_fix_simple_typo	False	File not fixed: This is a silly typo.	0	RuntimeError: Agent reached maximum iteration in headless mode. Current iteration: 30, max iteration: 30
		Really!
		No more typos.
		Enjoy!

---

Integration Tests Report Delegator (DeepSeek)
Success rate: 100.00% (2/2)

Total cost: USD 0.00

instance_id	success	reason	cost	error_message
t01_fix_simple_typo	True		0.001	nan
t02_add_bash_hello	True		0.001	nan

---

Integration Tests Report VisualBrowsing (DeepSeek)
Success rate: 100.00% (1/1)

Total cost: USD 0.00

instance_id	success	reason	cost	error_message
t05_simple_browsing	True		0	nan

---

Download testing outputs (includes both Haiku and DeepSeek results): Download

[enyst]

The DeepSeek API can take a LOOOONG time right now, locally too, I'll check what's up, but I don't think it has anything to do with this PR. 😅

The fix did eventually work locally.

[enyst]

This one is confusing the llm

[xingyaoww]

LGTM! And i agree with the comments robert has bring up -- thanks for fixing these :D