Databento-dbn

Enhancements
- Added `SymbolMappingMsgV2::new` method
- Added `Record` trait for all types beginning with a `RecordHeader`
- Added new `index_ts` and `raw_index_ts` methods to `Record` trait, which returns the
primary timestamp for a record
- Added `RecordMut` trait for accessing a mutable reference to a `RecordHeader`
- Implemented `PartialOrd` for all record types, based on `raw_index_ts`
- Loosened `DbnEncodable` from requiring `HasRType` to only requiring `Record`. This means
`RecordRef`s and concrete records can be encoded with the same methods

Breaking changes
- Split part of `HasRType` into new `Record` and `RecordMut` traits, which are object-
safe: they can be used in `Box<dyn>`. `RecordRef` also implements `Record`, so it's
easier to write code that works for both concrete records as well as `RecordRef`
- Removed `RecordRef` methods made redundant by it implementing `Record`
- Removed `input_compression` parameter from Python `Transcoder`

Deprecations
- Deprecated `SymbolIndex::get_for_rec_ref`, which was made redundant by loosening the
trait bound on `SymbolIndex::get_for_rec` to accept `RecordRef`s

Bug fixes
- Fixed `TsSymbolMap` not always using the correct timestamp for getting the mapped
symbol

Enhancements
- Added `map_symbols` support to Python `Transcoder`
- Added new publisher variants in preparation for DBEQ.PLUS dataset
- Added `from_dataset_venue` function to `Publisher` to facilitate destructuring
- Implemented `Default` for most records to make testing easier
- Added `from_zstd` function to `AsyncDbnEncoder` to match synchronous encoder
- Added re-exports for `enums::flags`, `enums::rtype`, `record::BidAskPair`,
`record::RecordHeader`, and `record::WithTsOut` to simplify imports
- Added `--fragment` CLI flag for writing DBN without the metadata header
- Added `--input-dbn-version` CLI option for specifying the DBN version of a DBN
fragment
- Added `serde::Deserialize` implementations for `Dataset`, `Venue`, and `Publisher`
- Added support for Python 3.12 to `databento_dbn`
- Added `RecordDecoder::with_version` for future use when dealing with compatibility
between different DBN versions
- Added new dispatch macros: `rtype_ts_out_method_dispatch`,
`rtype_ts_out_async_method_dispatch`, `rtype_method_dispatch`, and
`schema_ts_out_method_dispatch`
- Added `InstrumentDefMsgV2` and `SymbolMappingMsgV2` for forward compatibility with a
version of DBN
- Added `TsSymbolMap` and `PitSymbolMap` to aid with both historical and live symbology
- Added support for inverse symbology, i.e. with `stype_in=InstrumentId`

Breaking changes
- Changed `Metadata::symbol_map` to return `TsSymbolMap`
- Changed `Metadata::symbol_map_for_date` to return `PitSymbolMap`
- Changed `Default` implementation for `BidAskPair` by setting prices to `UNDEF_PRICE`
- Added new publisher values in preparation for DBEQ.PLUS
- Added `ts_out` parameter to `encode_header_for_schema` in `CsvEncoder` and
`DynEncoder` to allow controlling whether "ts_out" is in the header

Enhancements
- Upgraded `async-compression` to 0.4.3
- Upgraded `csv` to 1.3
- Upgraded `num_enum` to 0.7

Bug fixes
- Changed DBN stream detection to ignore the DBN version

Enhancements
- Added new `EncodeRecordTextExt` trait which is implemented for the CSV and JSON
encoders. It adds two methods for encoding a `symbol` field along side the rest of the
record fields, matching the behavior of `map_symbols` in the historical API
- Added `encode_header` and `encode_header_for_schema` methods to `CsvEncoder` and
`DynEncoder` to give more flexibility for encoding CSV headers
- Added `from_file` and `from_zstd_file` functions to `AsyncDbnDecoder` to match
synchronous decoder
- Implemented `Copy` for `RecordRef` to make it behave more like a reference
- Added `AsyncDbnEncoder` for simpler DBN encoding and to match sync API
- Added `RecordEnum` and `RecordRefEnum` to more easily be able to pattern match on
records of different types
- Added `ARCX.PILLAR.ARCX` publisher
- Added `From` DBN records for `RecordRef`
- Added re-exports to the top level of the crate for all enums and records for simpler
imports
- Added `ClosePrice` and `NetChange` `StatType`s used in the `OPRA.PILLAR` dataset

Breaking changes
- Split `encode_record_ref` into a safe method with no arguments and an unsafe method
with a `ts_out` parameter to reduce `unsafe` usage when not working with live data
that may contain `ts_out`

Bug fixes
- Fixed `dbn` CLI not writing CSV header when using `--fragment` and `--zstd-fragment`
flags
- Fixed lifetime on return value from `RecordRef::get_unchecked`
- Fixed missing check for `stype_out` before building `Metadata` symbology maps

Bug fixes
- Fixed query range checking in `Metadata::symbol_map_for_date`
- Added `debug_assert_eq!` check for alignment in `RecordRef::new`

Bug fixes
- Changed `Metadata::symbol_map` and `symbol_map_for_date` to return `String` values
instead of `&str`, which made it difficult to use