docs: consolidate RFCs + research into docs/rfcs/ + docs/research/

[aligatoralex]

Description of Changes:

Consolidates 7 separate docs-only draft PRs (#31, #33, #36, #38, #40, #42, #45), each on its own long-lived branch for a single markdown file, into one PR. This closes the gap between how this repo was accumulating design-proposal PRs/branches and how mature open-source projects actually do it.

Research behind this change: reviewed Rust RFCs (rust-lang/rfcs), Python PEPs (python/peps), Kubernetes KEPs (kubernetes/enhancements), and ADRs (adr.github.io). Common thread: none keep a proposal document on a long-lived, dedicated branch. A branch exists only long enough to open one PR; the document then lives as a file on main, and later changes (status updates, corrections) are small follow-up PRs to that same file — not new branches. This team already discusses in Issues rather than PR threads, which maps cleanly onto the ADR convention (discussion wherever the team prefers) rather than Rust's PR-thread-only model.

What this PR does:

• New docs/rfcs/ (plural) as the single RFC directory, superseding docs/rfc/ (singular, used by RFCs 0001-0004's now-closed PRs).
• docs/rfcs/0000-cfd-solver-strategy.md renumbered to 0005 for sequential continuity; all cross-references (including the RFC's own self-referential open question about this exact convention split) updated.
• Every RFC gets a Discussion: #NNN header pointing at its existing tracking issue ([RFC] Native solid-motor performance modeling (SolidMotorPerformanceAnalysis) #37, [RFC] Close ComponentStore gaps for physics-owned parameters (#28 follow-up) #39, [RFC] SUAVE dependency isolation (suave_compat shim + dedicated venv + CI job) #41, [RFC] CFD solver strategy — SU2 first-class, OpenFOAM optional, PyFluent excluded from public repo pending license confirmation #46) — issues stay the discussion venue, unchanged.
• RFC 0004 (OpenMDAO) explicitly notes no discussion issue was opened, respecting lead engineer @Arseni10Lk's ownership of that direction (FlightDeck/README.md) — this PR does not create a parallel tracking issue for his direction.
• docs/rfcs/README.md documents the adopted convention going forward (one file per RFC, status in header, short branch per change, discussion in Issues) and a table of current RFCs.
• docs/REPO_MAP.md and docs/SUAVE_INTEGRATION_NOTES.md move in unchanged (reference docs, not RFCs); docs/research/mdo-directions-2026.md holds the MDO-directions research digest.

Follow-up (next, after this merges): close PRs #31/#33/#36/#38/#40/#42/#45 with a comment pointing here, delete their branches, and update issues #37/#39/#41/#46 to link the new file paths instead of the old branches.

Checklist:

• I have run uv run pytest YAADO_Core/tests/ and all tests pass. (Docs-only; 102 passed.)
• My code includes type hints and docstrings. (N/A — Markdown.)
• I added all vital READMEs. (Adds docs/rfcs/README.md.)

Verification method:
Ran the full suite on this branch (102 passed, unaffected — no code touched). Cross-checked every docs/rfc/ path reference across all consolidated files and updated them; verified no other file references the old singular path after the edit.

---

Generated by Claude Code

[Arseni10Lk]

I had to clean the repo of the proprietary data from docs/ from the past. So the whole history was rebased

[aligatoralex]

Thanks! I was thinking about doing it myself - you read my mind :D

…

On 3 Sep 2026, at 13:25, Arseni Lysak ***@***.***> wrote: Arseni10Lk left a comment (knnmelprop/YAADO#47) <#47 (comment)> I had to clean the repo of the proprietary data from docs/ from the past. So the whole history was rebased — Reply to this email directly, view it on GitHub <#47?email_source=notifications&email_token=AF3KN5A45RVPG2NI3G7JUX35NFIL3A5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTKNJSGQ4TQMJQGE2KM4TFMFZW63VGMF2XI2DPOKSWK5TFNZ2KYZTPN52GK4S7MNWGSY3L#issuecomment-5524981014>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AF3KN5CYVYKOSQHGLH2TTAD5NFIL3AVCNFSNUABGKJSXA33TNF2G64TZHMYTEOJVG44DCNJVG45US43TOVSTWNJTGA3DGMRQGA4TJILWAI>. You are receiving this because you authored the thread.