* ⬆ [pre-commit.ci] pre-commit autoupdate. PR [#14237](https://github.com/fastapi/fastapi/pull/14237) by [@pre-commit-ci[bot]](https://github.com/apps/pre-commit-ci).
* ⬆ Bump actions/upload-artifact from 4 to 5. PR [#14235](https://github.com/fastapi/fastapi/pull/14235) by [@dependabot[bot]](https://github.com/apps/dependabot).

## 0.120.1

### Upgrades

        steps to [XLA](https://www.tensorflow.org/xla). Note that
        `jit_compile=True` may not necessarily work for all models.
    *   Added DTensor-related Keras APIs under `tf.keras.dtensor` namespace. The
        APIs are still classified as experimental. You are welcome to try it
        out. Please check the tutorial and guide on https://www.tensorflow.org/
        for more details about DTensor.

pkg errors, func Join(...error) error #53435
pkg fmt, func FormatString(State, int32) string #51668
pkg go/ast, type File struct, FileEnd token.Pos #53202
pkg go/ast, type File struct, FileStart token.Pos #53202
pkg go/ast, type RangeStmt struct, Range token.Pos #50429
pkg go/token, method (*FileSet) RemoveFile(*File) #53200
pkg go/types, func Satisfies(Type, *Interface) bool #56548
pkg io/fs, var SkipAll error #47209

> 8 { panic("NAF digits must fit in int8") } var naf [256]int8 var digits [5]uint64 for i := 0; i < 4; i++ { digits[i] = byteorder.LEUint64(b[i*8:]) } width := uint64(1 << w) windowMask := uint64(width - 1) pos := uint(0) carry := uint64(0) for pos < 256 { indexU64 := pos / 64 indexBit := pos % 64 var bitBuf uint64 if indexBit < 64-w { // This window's bits are contained in a single u64 bitBuf = digits[indexU64] >> indexBit } else { // Combine the current 64 bits with bits from the next 64 bitBuf...

> 8 { panic("NAF digits must fit in int8") } var naf [256]int8 var digits [5]uint64 for i := 0; i < 4; i++ { digits[i] = byteorder.LEUint64(b[i*8:]) } width := uint64(1 << w) windowMask := uint64(width - 1) pos := uint(0) carry := uint64(0) for pos < 256 { indexU64 := pos / 64 indexBit := pos % 64 var bitBuf uint64 if indexBit < 64-w { // This window's bits are contained in a single u64 bitBuf = digits[indexU64] >> indexBit } else { // Combine the current 64 bits with bits from the next 64 bitBuf...