[FEATURE] Expand available contexts, objects, and actions

[ncudlenco]

Is your feature request related to a problem? Please describe.
The current set of contexts, objects, and actions in mta-sim may be limited, restricting the richness of simulated scenarios, story generation, and downstream data annotation tasks. Expanding these elements will enhance the expressiveness, usability, and scalability of the simulation environment.

Describe the solution you'd like

• Identify gaps and limitations in the current lists of contexts (locations, situations), objects (interactable and decorative), and actions (atomic and composite).
• Expand the available contexts, objects, and actions to support a broader variety of scenarios and stories.
• Ensure new elements are well-documented and integrated into the simulation, annotation pipelines, and content generation tools.
• Where feasible, align additions with real-world datasets or standards (e.g., COCO, Visual Genome for objects; FrameNet for actions/context).

Describe alternatives you've considered

• Working with the current limited set (restricts research and application potential)
• Ad hoc additions without systematization (leads to inconsistencies and maintenance issues)

Acceptance Criteria

• Gaps in current contexts, objects, and actions are identified and documented
• Expanded lists of contexts, objects, and actions are developed and reviewed
• New elements are integrated into the simulation and related tools
• Documentation is updated for all new contexts, objects, and actions
• Tests verify the presence and functionality of new elements
• No existing functionality is broken
• Performance impact is acceptable

Additional context

• This feature will enable richer storytelling, better simulation coverage, and more comprehensive datasets.
• Can be coordinated with data annotation and story generation pipelines.