chore: Update faer requirement from 0.20.2 to 0.21.5

[dependabot]

Updates the requirements on faer to permit the latest version.

Changelog

Sourced from faer's changelog.

0.19

• Support matrix-scalar multiplication/division without the Scale wrapper for f32/f64.
• Implemented conjugate gradient, BiCGSTAB, and LSMR iterative solvers (currently gated by the unstable feature).
• Implemented Hermitian matrix pseudoinverse implementation. Thanks @​lishen_ for the contribution.
• Implemented column and row mean and variance in faer::stats.
• Added more iterator and parallel iterator functions (MatRef::[col|row]_partition, MatRef::par_[col|row]_partition, etc.).
• Added full and zeros constructors to owned Col, Row, and Matrix (issue-125).
• Added shape function to return both the row and the column count of a matrix.
• Added several missing associated functions from the mut and owning variants of matrices.
• Implemented core::iter::{Sum, Product} for c32 and c64.
• Sparse Cholesky can now be used with user-provided permutations.
• Simplified matrix constructors, adding a variant with a _generic prefix for the old behavior.
• LDLT and Bunch-Kaufman decompositions now stores the diagonal blocks instead of their inverses. This helps avoid infinities and NaNs when dealing with singular matrices.
• Integrated nano-gemm as a backend for small matrix multiplication.
• Significant performance improvements for small LLT and LDLT decompositions.

0.18

• Refactored the project so that faer contains all the core and decomposition implementations. faer-{core,cholesky,lu,qr,svd,evd,sparse} are now deprecated and will no longer be updated.
• Improved the multithreaded performance of the Eigenvalue decomposition for large matrices.
• Decomposition solve functions now accept column vectors as well as matrices.
• Implemented the L1 norm, and the squared L2 norm.
• Implemented conversions from sparse to dense matrices, by calling mat.to_dense().
• Sparse matrices now support duplicated entries. Note that faer will not add duplicated entries to a matrix unless the opposite is explicitly mentioned in the function documentation. faer also will deduplicate entries when created with Sparse{Col,Row}Mat::try_new_from_indices and other similar functions.
• Implemented conversions from unsorted to sorted sparse matrices by calling mat.to_sorted() (or mat.sort_indices() for owned matrices).
• Implemented {Col,Row}::try_as_slice[_mut] functions that return data as a slice if it is contiguous.
• Implemented .for_each_with_index and .map_with_index for the matrix zipping API, which passes the matrix row and column indices as well as the values.
• Added rand support for randomly generating matrices in the faer::stats module, as well as for faer::complex_native::{c32,c64}.
• Implemented a pseudoinverse helper for the high level SVD and thin SVD decompositions.

0.17

• Implemented sparse matrix arithmetic operators (other than sparse-sparse matrix multiplication), and added mutable sparse views as well as owning sparse matrix containers.
• Implemented try_from_triplets for sparse matrices.
• Re-exported subcrates in faer::modules.
• Improved performance of the SVD decomposition for small matrices.
• Implemented col!, row! and concat! macros. Thanks @​DeliciousHair for the contribution.
• Implemented more c32/c64 operations. Thanks @​edyounis for the contribution.
• Implemented the Kronecker product in faer_core. Thanks @​edyounis for the contribution.
• Implemented (de)serialization of Mat. Thanks @​cramt for the contribution.

0.16

• Implemented the index operator for row and column structures. Thanks @​DeliciousHair for the contribution.
• Exposed a few sparse matrix operations in the high level API.
• Implemented sparse LU and QR, and exposed sparse decompositions in the high level API.
• Better assertion error messages in no_std mode.

0.15

• Implemented initial API of Row/RowRef/RowMut and Col/ColRef/ColMut structs for handling matrices with a single row or column.
• Implemented [Mat|Col|Row]::norm_l2 and [Mat|Col|Row]::norm_max for computing the L2 norm of a matrix or its maximum absolute value.
• Fixed several bugs in the eigenvalue decompositions. Special thanks to @​AlexMath for tracking down the errors.
• Updated zipped! macro API, which now requires a matching unzipped! for matching the closure arguments.

... (truncated)

Commits

• e6ec990 chore: Release
• 1fc8cd6 add test case
• 5dc1dff fix matrix reserve + add nano_gemm
• 3b89fbc fix matrix resize
• 06600ce chore: Release
• 7bc1567 fix sum and activate v3 feature in pulp
• 56028f4 minor fixes, apply Jed's paper changes
• 39e72ec fix formatting
• 16d6d0e adds as_ref to SparseRowMat
• ed9dcf0 adds as_ref to SparseColMat
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot merge will merge this PR after your CI passes on it
• @dependabot squash and merge will squash and merge this PR after your CI passes on it
• @dependabot cancel merge will cancel a previously requested merge and block automerging
• @dependabot reopen will reopen this PR if it is closed
• @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

[dependabot]

Dependabot encountered an unknown error. Because of this, Dependabot cannot update this pull request.

[dependabot]

OK, I won't notify you again about this release, but will get in touch when a new version is available. If you'd rather skip all updates until the next major or minor version, let me know by commenting @dependabot ignore this major version or @dependabot ignore this minor version. You can also ignore all major, minor, or patch releases for a dependency by adding an ignore condition with the desired update_types to your config file.

If you change your mind, just re-open this PR and I'll resolve any conflicts on it.