v0.2.0

What's Changed

• Phase 1: Foundation + Discover page by @neuromechanist in #7
• Phase 2: Dataset detail page + on* provenance + version switcher by @neuromechanist in #8
• Phase 2 polish: 6 design-review fixes (fonts, a11y, tree layout, keywords) by @neuromechanist in #9
• fix(provenance): prefer canonical ds id when multiple IsDerivedFrom entries by @neuromechanist in #20
• feat: 301 redirect ds*->on* on dataset detail page by @neuromechanist in #18
• perf: edge-cache via caches.default in middleware by @neuromechanist in #14
• feat: NEMAR-branded favicon, apple+android icons, web manifest by @neuromechanist in #15
• feat: OG/Twitter/canonical meta + brand OG card by @neuromechanist in #16
• feat: prefer authors over uploader in DatasetCard byline by @neuromechanist in #23
• phase 3: QualityPanel with runtime aggregator from per-file dataqual by @neuromechanist in #24
• Redesign nemar.org: Phases 1+2+3a on Cloudflare Pages + Astro by @neuromechanist in #26
• cleanup: redact account id, drop orphan, log silent catches by @neuromechanist in #27
• feat(dataset): short-circuit unpublished datasets in readme + tree endpoints by @neuromechanist in #30
• feat(dataset): consume compact summary.json with manifest fallback by @neuromechanist in #31
• refactor(data-api): tag jsonFetch outcomes instead of collapsing to null by @neuromechanist in #33
• Phase 1: Passwordless auth + cookie notice + remember-me by @neuromechanist in #39
• Phase 2: Dataset upload flow by @neuromechanist in #40
• Phase 3: My datasets dashboard by @neuromechanist in #41
• Phase 4: Collaborator + publication review by @neuromechanist in #42
• Phase 5: Onboarding, docs, and cutover by @neuromechanist in #43
• Researcher dashboard: sign up, upload, manage your datasets by @neuromechanist in #45
• Split nemar.org and app.nemar.org surfaces (#47) by @neuromechanist in #48
• Route cross-host links to ww2.nemar.org (#49) by @neuromechanist in #50
• Fix login verify: forward Origin + error UX (#51) by @neuromechanist in #52
• Accept backend /auth/me shape (id:number, role:member) (#55) by @neuromechanist in #56
• Accept 'owner' role + fix support email (#57) by @neuromechanist in #58
• fix(api): no-store on tree/readme empty-fallback branches by @neuromechanist in #54
• Same-origin /api/v1 proxy for dashboard mutations (#59) by @neuromechanist in #60
• Hotfix: /api/v1 is an app route, not marketing by @neuromechanist in #61
• Hotfix: move proxy test out of src/pages (Pages build fail) by @neuromechanist in #62
• Phase 1 (#65): bump edge-cache namespace to v2 by @neuromechanist in #67
• Phase 2 (#63): consume summary.readme.content from schema 1.1 by @neuromechanist in #68
• Phase 3 (#64): progressive BIDS tree (skeleton + lazy per-subject) by @neuromechanist in #69
• Epic #66: Cold-paint perf + cache hygiene for dataset detail by @neuromechanist in #70
• #71: Consume nemar-cli //page-bundle.json on dataset detail by @neuromechanist in #72
• #74: Tree totals from summary.totals.bytes + 4s summary timeout on partials by @neuromechanist in #75
• #76: Pivot dataset detail to canonical sources (drop summary/bundle/partials) by @neuromechanist in #77
• Lazy chunked dir tree + drop catalog-row SSR fetch by @neuromechanist in #79
• Drop QA SSR; Demographics tab (live) + Data quality tab (coming soon) by @neuromechanist in #82
• Demographics chart fixes: tick sizing, fixed height, null/Other, coming-soon tab by @neuromechanist in #84
• File-row inline viewers (md/json/tsv) + DatasetTabs pills + design language doc by @neuromechanist in #86
• Mobile layout: strip DOI badge, overflow guards, stacked rail / version switcher by @neuromechanist in #88
• Detail page: dedupe description vs README, more file-row padding by @neuromechanist in #89
• Detail page: keep description, match README prose, add AI-generated marker by @neuromechanist in #90
• Detail page: description as a card matching the README by @neuromechanist in #91
• Mobile fix: grid track sizing — minmax(0, 1fr) so the file tree's min-content doesn't widen the page by @neuromechanist in #92
• Consolidate tag system: one clickable Tag element + color bible by @neuromechanist in #93
• feat(discover): hybrid lexical+semantic search (closes #12) by @neuromechanist in #95
• fix(discover): license filter works (server-side) (#96) by @neuromechanist in #97
• fix(filters): multi-select license/modality (form repeated params) (#98) by @neuromechanist in #100
• fix(rail): render keywords as full-width block to prevent pill overflow by @neuromechanist in #102
• feat: interactive Zarr signal viewer (website#99) by @neuromechanist in #101
• feat: viewer P1 - filters, full-res lines, butterfly, minimap, cursor, keyboard (website#99) by @neuromechanist in #103
• feat(viewer): gear settings menu + notch default from PowerLineFrequency + cursor white-flash fix by @neuromechanist in #104
• feat(viewer): cursor overlay, proportionate fonts, prominent minimap events, 45deg labels, event descriptions by @neuromechanist in #105
• feat(viewer): bad-channel reject/hide mode (P2) by @neuromechanist in #106
• feat(viewer): scalp topomap split-panel (EEGLAB-style, viridis) — P2 by @neuromechanist in #107
• feat(viewer): WebGL2 trace rendering (P2) by @neuromechanist in #108
• Point Documentation link to docs.nemar.org by @neuromechanist in #110
• Retire in-site /docs + citation-dashboard; redirect to subdomains by @neuromechanist in #111
• Flag web sign-in as coming soon (July 2026) by @neuromechanist in #112
• Speed up Zarr viewer startup by @neuromechanist in #113
• Add dataset OG images by @neuromechanist in #114
• Improve dataset OG cards by @neuromechanist in #115
• Landing redesign: legacy dark hero, hosted-only stats, footer credits by @neuromechanist in #117
• Viewer: explain why a recording has no interactive view by @neuromechanist in #116
• Fix dataset OG PNG previews by @neuromechanist in #119
• Use static dataset OG PNGs by @neuromechanist in #120
• Version dataset OG image URLs by @neuromechanist in #121
• Fix dataset OG card text by @neuromechanist in #122
• Add scheduled OG rebuild by @neuromechanist in #123
• feat(viewer): standard-montage topomap fallback by @neuromechanist in #118
• dataset: wget/curl recipe for skipped (large) archives (#752) by @neuromechanist in #124
• Add fNIRS and Motion modalities to filter and color scheme by @neuromechanist in #125
• feat: citations pill + modal, listing count, sort by @neuromechanist in #127
• feat: ORCID SSO frontend (login button + OAuth proxy + onboarding) by @neuromechanist in #130
• feat: collect city/country/affiliation on ORCID signup completion by @neuromechanist in #131
• feat(viewer): EGI/BioSemi named-net topomap fallback (nemar-cli#855) by @neuromechanist in #136
• feat(discover): activate channel density + montage filter by @neuromechanist in #137
  *...