chore(lint): enforce ruff pyupgrade (UP) rules (ga4gh#562)

close ga4gh#561

fix-logging-warnings (ga4gh#552)

This PR stops logging unnecessary warnings like:
```
Derived sequence '' is different from provided state.sequence 'root='''

Derived sequence 'CCCC' is different from provided state.sequence 'root='CCCC''
```

An example for reproducing the warning log. Functionality is correct,
but the condition for logging has a small bug.

```python
>>> from ga4gh.vrs.extras.translator import AlleleTranslator
>>> from ga4gh.vrs.dataproxy import create_dataproxy
>>> dp = create_dataproxy('seqrepo+file:///usr/local/share/seqrepo/2024-12-20')
>>> translator = AlleleTranslator(data_proxy=dp)
>>> allele = translator.translate_from('21-5030497-C-CACCT', 'gnomad')
>>> translator.translate_to(allele, 'spdi')
Derived sequence 'ACCTACCT' is different from provided state.sequence 'root='ACCTACCT''
['NC_000021.9:5030497:4:ACCTACCT']
```

Fix allele translator to_spdi issue for RLE deletion. Add RLE reconst…

…ruction when longer than rle_seq_limit. (ga4gh#542)

This fixes a bug for `_to_spdi` for ReferenceLengthExpressions. 

It also adds an optional argument to `translate_to` called
`ref_seq_limit` which works similarly to `rle_seq_limit`, but for SPDI
expressions. If `ref_seq_limit` is nonzero, the 3rd term of the SPDI
expression will be the actual reference sequence, not the length.

To round-trip a clinvar SPDI expression which contains the reference
sequence:

```
spdi = "NC_000019.10:44908821:C:T"
vrs = translator.translate_from(spdi, "spdi")
to_spdi = translator.translate_to(vrs, "spdi", ref_seq_limit=None) # or some number >=1 since len("C")==1
assert spdi == to_spdi
```

To continue the existing behavior of using the reference sequence
length:
```
spdi = "NC_000019.10:44908821:1:T"
vrs = translator.translate_from(spdi, "spdi")
to_spdi = translator.translate_to(vrs, "spdi") # ref_seq_limit defaults to 0
assert spdi == to_spdi
```

---------

Co-authored-by: Kori Kuzma <korikuzma@gmail.com>

feat: add optional suppression of allele collection (ga4gh#545)

* VCF annotator should not collect alleles if no output is requested by
user. This has major implications for memory usage, apparently.
* Provide this in a relatively OOP-friendly way without producing a
breaking change

At the next major version, there is a cleaner way to do this and it
should be implemented at that point

cicd: do not install setuptools (ga4gh#539)

We were missing an issue upstream about a deprecated stdlib module
because `setuptools` comes with an in-place fix -- but `setuptools`
wasn't an explicit dependency declared for downstream users. I think we
should just start with whatever `setup-python` gives us unless there's a
pressing need.

chore: remove unused pylint config (ga4gh#525)

fix(ci): revert ga4gh#508 and ga4gh#509 + exclude HEAD from git branc…

…h output in pypi release job (ga4gh#510)

feat!: update models to vrs 2.0.0 community review ballot (ga4gh#471)

close ga4gh#469 

* Update modules to vrs
[2.0.0-ballot.2024-11.3](https://github.com/ga4gh/vrs/tree/2.0.0-ballot.2024-11.3)
tag
* Rename class `_Ga4ghIdentifiableObject` to `Ga4ghIdentifiableObject`
* Removes `entity_models` and `domain_models` from `ga4gh.core`. Now,
all gks-core models are in `ga4gh.core.models`.
* gks-core includes `date` and `datetime` as primitives, however the
builtin `datetime` module already provides these. Initially, I kept them
as Pydantic Root Models but I think we should just leverage the
`datetime` module instead.

feat!: update gks-common / vrs models (ga4gh#453)

close ga4gh#452

docs: fix param typo in vcf annotator (ga4gh#451)