webui: auto-refresh /props on inference start to resync model metadata

[ServeurpersoCom]

• Add no-cache headers to /props and /slots

• Throttle slot checks to 30s

• Prevent concurrent fetches with promise guard

• Trigger refresh from chat streaming for legacy and ModelSelector

• Show dynamic serverWarning when using cached data

• Updated assistant message bubbles to show each message's stored model when available,
  falling back to the current server model only when the per-message value is missing

• When the model selector is disabled, now fetches /props and prioritizes that model name
  over chunk metadata, then persists it with the streamed message so legacy mode properly
  reflects the backend configuration

Cmdline used on legacy (Raspberry Pi 5) :

/root/llama.cpp/build/bin/llama-server \
 -m /root/ia/models/mradermacher/OLMoE-1B-7B-0125-Instruct-i1-GGUF/OLMoE-1B-7B-0125-Instruct.i1-Q6_K.gguf \
 -ctk q8_0 -ctv q8_0 -fa on \
 --jinja --ctx-size 8192 --mlock --port 8081

/root/llama.cpp/build/bin/llama-server \
 -m /root/ia/models/mradermacher/OLMoE-1B-7B-0125-SFT-i1-GGUF/OLMoE-1B-7B-0125-SFT.i1-Q6_K.gguf \
 -ctk q8_0 -ctv q8_0 -fa on \
 --jinja --ctx-size 8192 --mlock --port 8081

/root/llama.cpp/build/bin/llama-server \
 -m /root/ia/models/mradermacher/Qwen3-30B-A3B-Instruct-2507-i1-GGUF/Qwen3-30B-A3B-Instruct-2507.i1-Q4_K_M.gguf \
 --temp 0.7 --top-p 0.8 --top-k 20 --min-p 0 \
 -ctk q8_0 -ctv q8_0 -fa on \
 --jinja --ctx-size 4096 --port 8081

Fixes #16771

EDIT : I've recorded a video that specifically targets the original issue.
Testing video, Raspberry Pi 5 + master branch + this PR :

CmdLineSwap-RaspberryPi5.mp4

And another one to show there's no regression when the model selector is enabled,
also demonstrating the multimodal function updates:

NonReg-FullSetup.mp4

[ServeurpersoCom]

The final fix is done, now that the /props refresh is unified, the legacy mode (without the model selector) behaves correctly.
The message history now properly reflects the backend model used.

[ServeurpersoCom]

LEGACY MODE (modelSelectorEnabled = false)

• Do not send "model" in requests
• Ignore "model" from server responses
• Read /props.model as the single source of truth
• Refresh /props periodically or on stream start
• UI and history reflect the backend model

MODEL SELECTOR MODE (modelSelectorEnabled = true)

• Always send "model" in requests
• Use per-message "model" from chunks or user selection
• /props used only for backend capabilities (context, multimodal, slots)
• UI reflects the user-selected model

Summary:
Legacy -> backend decides the model via /props
ModelSelector -> frontend decides the model via request field

[allozaur]

Yes, @ServeurpersoCom, that is much better than the initial idea for this PR 😄

[ServeurpersoCom]

Yes, @ServeurpersoCom, that is much better than the initial idea for this PR 😄

Good thing the merge stashes everything, those intermediate commits were an emotional rollercoaster 😆 Glad we made it out alive 😆

[allozaur]

those intermediate commits were an emotional rollercoaster 😆

🤣

[ServeurpersoCom]

rebased to retry CI

[ServeurpersoCom]

Rebased + static build