Add soft evaluate validation and whitespace-tolerant quotes

Fix live /v2/evaluate judge and trainer failures on LocalAI

chore: update submodules

Merge pull request #14 from xyz-nebula/feat/server-negotiation-results

debug

docs: describe backend evaluation results

feat: preserve mock result profiles

feat: adapt server result screen

debug

Merge pull request #13 from xyz-nebula/user-manual

feat: document product tour architecture, routes, and testing specifications

feat: enhance product tour with beacon and tooltip improvements

feat: enhance product tour with error handling and recovery features

chore: update submodules

docs: describe backend evaluation results

feat: preserve mock result profiles

feat: adapt server result screen

feat: add deadline-based result polling

feat: connect backend evaluation workflow

Merge pull request #13 from xyz-nebula/user-manual

feat: document product tour architecture, routes, and testing specifications

feat: enhance product tour with beacon and tooltip improvements

feat: enhance product tour with error handling and recovery features

feat: add product tour joyride UI

Merge pull request #12 from xyz-nebula/dev

docker: change args

Merge pull request #11 from xyz-nebula/docker

package update

feat: add Docker build and GHCR publish workflow

feat: document product tour architecture, routes, and testing specifications

feat: enhance product tour with beacon and tooltip improvements

feat: enhance product tour with error handling and recovery features

feat: add product tour joyride UI

feat: connect product tour targets and events

feat: add product tour invitation and state machine

chore: update submodules

Improve v2 evaluation diagnostics and judge evidence checks

feat: monorepo with submodules tracking dev

updated README