feat(storage): add model metadata tables

[tanishqgandhi1908]

What changes were proposed in this PR?

Adds the database tables that track ML models as a first-class resource in Texera. The model* tables mirror the existing dataset* tables — their own primary key (mid), their own LakeFS repo namespace — and
add model-specific attributes (framework, format):

• model — the model asset (owner, name, visibility, description, cover image, framework, format)
• model_user_access — per-user privilege (reuses privilege_enum)
• model_version — versions of a model (mvid, version_hash)
• model_upload_session / model_upload_session_part — multipart upload bookkeeping
• model_user_likes / model_view_count — like/view counters

The change is pure schema, with no application code and no behavior change:

• Fresh installs: tables added to sql/texera_ddl.sql (with matching DROP TABLE entries and FK-safe ordering).
• Existing deployments: a new idempotent migration sql/updates/30.sql (all CREATE TABLE IF NOT EXISTS, no data backfill), registered as changeSet id="30" in sql/changelog.xml.
• The JOOQ data-access layer is git-ignored and regenerated from the live DB on build, so no generated code is committed.

Migration ordering note: this uses id=30 to sit after #6495's migration 29 (PR #6502, currently in review). These model tables do not depend on #6495, so this PR can be reviewed independently.

Any related issues, documentation, discussions?

Closes #6496.

How was this PR tested?

Validated against a throwaway groonga/pgroonga:4.0.1-debian-15 Postgres:

• sql/texera_ddl.sql runs end-to-end with no errors; all 7 model* tables are created with the expected primary keys, foreign keys (ON DELETE CASCADE), unique constraints, and CHECK constraints.
• sql/updates/30.sql applies cleanly on a DB that already has the dataset tables but no model tables, and is idempotent

Was this PR authored or co-authored using generative AI tooling?

Generated-by: Claude Code (Claude Opus 4.8)

[github-actions]

Automated Reviewer Suggestions

Based on the git blame history of the changed files, we recommend the following reviewers:

• Contributors with relevant context: @kunwp1
  You can notify them by mentioning @kunwp1 in a comment.

[github-actions]

⚠️ Benchmark changes need a look

🟢 0 better · 🔴 2 worse · ⚪ 13 noise (<±5%) · 0 without baseline

Compared against main a9f384e benchmarked on this same runner, so the delta is largely free of cross-runner hardware noise. The "7d avg" column still reflects the gh-pages dashboard. Treat <±5% as noise unless repeated.

Dashboard · Run

	config	throughput	MB/s	latency	max Δ latest / 7d
⚪	bs=10 sw=10 sl=64	419	0.256	23,401/31,031/31,031 us	⚪ within ±5% / 🔴 +92.5%
🔴	bs=100 sw=10 sl=64	927	0.566	105,383/155,687/155,687 us	🔴 +16.5% / 🔴 +44.9%
⚪	bs=1000 sw=10 sl=64	1,127	0.688	885,944/918,819/918,819 us	⚪ within ±5% / 🟢 -13.0%

Baseline details

Latest main a9f384e from same runner

config	metric	PR	latest main	7d avg	Δ latest	Δ 7d
bs=10 sw=10 sl=64	throughput	419 tuples/sec	439 tuples/sec	760.2 tuples/sec	-4.6%	-44.9%
bs=10 sw=10 sl=64	MB/s	0.256 MB/s	0.268 MB/s	0.464 MB/s	-4.5%	-44.8%
bs=10 sw=10 sl=64	p50	23,401 us	22,596 us	12,781 us	+3.6%	+83.1%
bs=10 sw=10 sl=64	p95	31,031 us	31,851 us	16,118 us	-2.6%	+92.5%
bs=10 sw=10 sl=64	p99	31,031 us	31,851 us	19,074 us	-2.6%	+62.7%
bs=100 sw=10 sl=64	throughput	927 tuples/sec	960 tuples/sec	988.68 tuples/sec	-3.4%	-6.2%
bs=100 sw=10 sl=64	MB/s	0.566 MB/s	0.586 MB/s	0.603 MB/s	-3.4%	-6.2%
bs=100 sw=10 sl=64	p50	105,383 us	100,768 us	101,448 us	+4.6%	+3.9%
bs=100 sw=10 sl=64	p95	155,687 us	133,606 us	107,410 us	+16.5%	+44.9%
bs=100 sw=10 sl=64	p99	155,687 us	133,606 us	120,258 us	+16.5%	+29.5%
bs=1000 sw=10 sl=64	throughput	1,127 tuples/sec	1,122 tuples/sec	1,023 tuples/sec	+0.4%	+10.2%
bs=1000 sw=10 sl=64	MB/s	0.688 MB/s	0.685 MB/s	0.624 MB/s	+0.4%	+10.2%
bs=1000 sw=10 sl=64	p50	885,944 us	888,810 us	983,888 us	-0.3%	-10.0%
bs=1000 sw=10 sl=64	p95	918,819 us	930,691 us	1,030,100 us	-1.3%	-10.8%
bs=1000 sw=10 sl=64	p99	918,819 us	930,691 us	1,056,513 us	-1.3%	-13.0%

Raw CSV

config_idx,batch_size,schema_width,string_len,num_batches,total_ms,total_tuples,total_bytes,tuples_per_sec,mb_per_sec,lat_p50_us,lat_p95_us,lat_p99_us
0,10,10,64,20,476.80,200,128000,419,0.256,23401.03,31030.57,31030.57
1,100,10,64,20,2157.09,2000,1280000,927,0.566,105383.05,155687.17,155687.17
2,1000,10,64,20,17739.14,20000,12800000,1127,0.688,885944.17,918818.90,918818.90

[tanishqgandhi1908]

/request-review @aicam

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 75.20%. Comparing base (a9f384e) to head (de806ba).
⚠️ Report is 7 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #6799      +/-   ##
============================================
- Coverage     75.20%   75.20%   -0.01%     
+ Complexity     3447     3446       -1     
============================================
  Files          1160     1160              
  Lines         45807    45807              
  Branches       5071     5071              
============================================
- Hits          34451    34450       -1     
  Misses         9745     9745              
- Partials       1611     1612       +1     

Flag	Coverage Δ		*Carryforward flag
access-control-service	70.00% <ø> (ø)
agent-service	76.76% <ø> (ø)		Carriedforward from a9f384e
amber	67.17% <ø> (-0.01%)	⬇️
computing-unit-managing-service	20.49% <ø> (ø)
config-service	66.66% <ø> (ø)
file-service	67.21% <ø> (ø)
frontend	79.42% <ø> (ø)		Carriedforward from a9f384e
notebook-migration-service	78.94% <ø> (ø)
pyamber	92.15% <ø> (ø)		Carriedforward from a9f384e
workflow-compiling-service	55.14% <ø> (ø)

*This pull request uses carry forward flags. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[aicam]

This PR is atomic but does not include any logic, please add more code to this PR to show the minimum way it will be used, its not possible to review a schema change without checking its effect on logics.