feat(hpu): Add Hpu backend implementation

This backend abstract communication with Hpu Fpga hardware.
It define it's proper entities to prevent circular dependencies with
tfhe-rs.
Object lifetime is handle through Arc<Mutex<T>> wrapper, and enforce
that all objects currently alive in Hpu Hw are also kept valid on the
host side.

It contains the second version of HPU instruction set (HIS_V2.0):
* DOp have following properties:
  + Template as first class citizen
  + Support of Immediate template
  + Direct parser and conversion between Asm/Hex
  + Replace deku (and it's associated endianess limitation) by
  + bitfield_struct and manual parsing

* IOp have following properties:
  + Support various number of Destination
  + Support various number of Sources
  + Support various number of Immediat values
  + Support of multiple bitwidth (Not implemented yet in the Fpga
    firmware)

Details could be view in `backends/tfhe-hpu-backend/Readme.md`

mockups/tfhe-hpu-mockup/params/gaussian_44b.toml

@@ -0,0 +1,44 @@
+[pbs_params]
+    lwe_dimension=724
+    glwe_dimension=2
+    polynomial_size=1024
+    lwe_noise_distribution={GaussianStdDev=1.2597809688976277e-05}
+    glwe_noise_distribution={GaussianStdDev=2.2737367544323206e-13}
+    pbs_base_log= 20
+    pbs_level= 1
+    ks_base_log= 2
+    ks_level= 7
+    message_width= 2
+    carry_width= 2
+    ciphertext_width= 44
+
+[ntt_params]
+    core_arch="WmmUnfoldPcg"
+    min_pbs_nb= 10
+    batch_pbs_nb= 16
+    total_pbs_nb= 32
+    ct_width= 44
+    radix= 2
+    stg_nb= 10
+    prime_modulus="Solinas2_44_14"
+    psi= 32
+    delta= 5
+[ks_params]
+    width= 21
+    lbx= 2
+    lby= 32
+    lbz= 3
+[pc_params]
+    ksk_pc= 4
+    ksk_bytes_w= 64
+    bsk_pc= 4
+    bsk_bytes_w= 64
+    pem_pc= 2
+    pem_bytes_w= 64
+    glwe_bytes_w= 64
+[regf_params]
+    reg_nb= 64
+    coef_nb= 32
+[isc_params]
+    min_iop_size= 4
+    depth= 64

mockups/tfhe-hpu-mockup/params/gaussian_44b_fast.toml

@@ -0,0 +1,44 @@
+[pbs_params]
+    lwe_dimension=20
+    glwe_dimension=2
+    polynomial_size=1024
+    lwe_noise_distribution={GaussianStdDev=0.0}
+    glwe_noise_distribution={GaussianStdDev=0.0}
+    pbs_base_log=20
+    pbs_level=1
+    ks_base_log=2
+    ks_level=7
+    message_width=2
+    carry_width=2
+    ciphertext_width=44
+
+[ntt_params]
+    core_arch="WmmCompactPcg"
+    min_pbs_nb= 10
+    batch_pbs_nb= 16
+    total_pbs_nb= 32
+    ct_width= 44
+    radix= 2
+    stg_nb= 10
+    prime_modulus="Solinas2_44_14"
+    psi= 32
+    delta= 5
+[ks_params]
+    width= 21
+    lbx= 2
+    lby= 32
+    lbz= 3
+[pc_params]
+    ksk_pc= 4
+    ksk_bytes_w= 64
+    bsk_pc= 4
+    bsk_bytes_w= 64
+    pem_pc= 2
+    pem_bytes_w= 64
+    glwe_bytes_w= 64
+[regf_params]
+    reg_nb= 64
+    coef_nb= 32
+[isc_params]
+    min_iop_size= 4
+    depth= 64

mockups/tfhe-hpu-mockup/params/gaussian_64b.toml

@@ -0,0 +1,47 @@
+[pbs_params]
+    lwe_dimension=724
+    glwe_dimension=2
+    polynomial_size=1024
+    lwe_noise_distribution={GaussianStdDev= 1.2597809688976277e-05}
+    glwe_noise_distribution={GaussianStdDev= 2.2737367544323206e-13}
+    pbs_base_log= 20
+    pbs_level= 1
+    ks_base_log= 2
+    ks_level= 7
+    message_width= 2
+    carry_width= 2
+    ciphertext_width= 64
+
+[ntt_params]
+    core_arch= {GF64=[5,5]}
+    min_pbs_nb= 10
+    batch_pbs_nb= 12
+    total_pbs_nb= 32
+    ct_width= 64
+    radix= 2
+    stg_nb= 10
+    prime_modulus= "GF64"
+    psi= 32
+    delta= 5
+
+[ks_params]
+    width= 21
+    lbx= 2
+    lby= 32
+    lbz= 3
+
+[pc_params]
+    ksk_pc= 4
+    ksk_bytes_w= 32
+    bsk_pc= 4
+    bsk_bytes_w= 32
+    pem_pc= 2
+    pem_bytes_w= 32
+    glwe_bytes_w= 32
+
+[regf_params]
+    reg_nb= 64
+    coef_nb= 32
+[isc_params]
+    min_iop_size= 4
+    depth= 64

mockups/tfhe-hpu-mockup/params/gaussian_64b_fast.toml

@@ -0,0 +1,44 @@
+[pbs_params]
+    lwe_dimension=20
+    glwe_dimension=2
+    polynomial_size=1024
+    lwe_noise_distribution={GaussianStdDev=0.0}
+    glwe_noise_distribution={GaussianStdDev=0.0}
+    pbs_base_log=20
+    pbs_level=1
+    ks_base_log=2
+    ks_level=7
+    message_width=2
+    carry_width=2
+    ciphertext_width=64
+
+[ntt_params]
+    core_arch= {GF64=[5,5]}
+    min_pbs_nb= 10
+    batch_pbs_nb= 12
+    total_pbs_nb= 32
+    ct_width= 64
+    radix= 2
+    stg_nb= 10
+    prime_modulus="GF64"
+    psi= 32
+    delta= 5
+[ks_params]
+    width= 21
+    lbx= 2
+    lby= 32
+    lbz= 3
+[pc_params]
+    ksk_pc= 4
+    ksk_bytes_w= 32
+    bsk_pc= 4
+    bsk_bytes_w= 32
+    pem_pc= 2
+    pem_bytes_w= 32
+    glwe_bytes_w= 32
+[regf_params]
+    reg_nb= 64
+    coef_nb= 32
+[isc_params]
+    min_iop_size= 4
+    depth= 64

mockups/tfhe-hpu-mockup/params/gaussian_64b_pfail64.toml

@@ -0,0 +1,47 @@
+[pbs_params]
+    lwe_dimension=804
+    glwe_dimension=1
+    polynomial_size=2048
+    lwe_noise_distribution={GaussianStdDev= 5.963599673924788e-6}
+    glwe_noise_distribution={GaussianStdDev= 2.8452674713391114e-15}
+    pbs_base_log= 23
+    pbs_level= 1
+    ks_base_log= 2
+    ks_level= 8
+    message_width= 2
+    carry_width= 2
+    ciphertext_width= 64
+
+[ntt_params]
+    core_arch= {GF64=[5,6]}
+    min_pbs_nb= 6
+    batch_pbs_nb= 12
+    total_pbs_nb= 32
+    ct_width= 64
+    radix= 2
+    stg_nb= 11
+    prime_modulus= "GF64"
+    psi= 32
+    delta= 5
+
+[ks_params]
+    width= 21
+    lbx= 3
+    lby= 64
+    lbz= 3
+
+[pc_params]
+    ksk_pc= 16
+    ksk_bytes_w= 32
+    bsk_pc= 8
+    bsk_bytes_w= 32
+    pem_pc= 2
+    pem_bytes_w= 32
+    glwe_bytes_w= 32
+
+[regf_params]
+    reg_nb= 64
+    coef_nb= 32
+[isc_params]
+    min_iop_size= 4
+    depth= 64

mockups/tfhe-hpu-mockup/params/gaussian_64b_pfail64_psi64.toml

@@ -0,0 +1,48 @@
+[pbs_params]
+    lwe_dimension=804
+    glwe_dimension=1
+    polynomial_size=2048
+    lwe_noise_distribution={GaussianStdDev= 5.963599673924788e-6}
+    glwe_noise_distribution={GaussianStdDev= 2.8452674713391114e-15}
+    pbs_base_log= 23
+    pbs_level= 1
+    ks_base_log= 2
+    ks_level= 8
+    message_width= 2
+    carry_width= 2
+    ciphertext_width= 64
+    opportunistic=true
+
+[ntt_params]
+    core_arch= {GF64=[5,6]}
+    min_pbs_nb= 11
+    batch_pbs_nb= 12
+    total_pbs_nb= 32
+    ct_width= 64
+    radix= 2
+    stg_nb= 11
+    prime_modulus= "GF64"
+    psi= 64
+    delta= 5
+
+[ks_params]
+    width= 21
+    lbx= 3
+    lby= 64
+    lbz= 3
+
+[pc_params]
+    ksk_pc= 16
+    ksk_bytes_w= 32
+    bsk_pc= 8
+    bsk_bytes_w= 32
+    pem_pc= 2
+    pem_bytes_w= 32
+    glwe_bytes_w= 32
+
+[regf_params]
+    reg_nb= 64
+    coef_nb= 32
+[isc_params]
+    min_iop_size= 4
+    depth= 64

mockups/tfhe-hpu-mockup/params/tuniform_64b_fast.toml

@@ -0,0 +1,44 @@
+[pbs_params]
+    lwe_dimension=20
+    glwe_dimension=1
+    polynomial_size=2048
+    lwe_noise_distribution={TUniformBound= 0}
+    glwe_noise_distribution={TUniformBound= 0}
+    pbs_base_log=23
+    pbs_level=1
+    ks_base_log=2
+    ks_level=7
+    message_width=2
+    carry_width=2
+    ciphertext_width=64
+
+[ntt_params]
+    core_arch= {GF64=[5,6]}
+    min_pbs_nb= 10
+    batch_pbs_nb= 12
+    total_pbs_nb= 32
+    ct_width= 64
+    radix= 2
+    stg_nb= 11
+    prime_modulus="GF64"
+    psi= 32
+    delta= 5
+[ks_params]
+    width= 21
+    lbx= 2
+    lby= 32
+    lbz= 3
+[pc_params]
+    ksk_pc= 4
+    ksk_bytes_w= 32
+    bsk_pc= 4
+    bsk_bytes_w= 32
+    pem_pc= 2
+    pem_bytes_w= 32
+    glwe_bytes_w= 32
+[regf_params]
+    reg_nb= 64
+    coef_nb= 32
+[isc_params]
+    min_iop_size= 4
+    depth= 64

mockups/tfhe-hpu-mockup/params/tuniform_64b_pfail64_psi64.toml

@@ -0,0 +1,48 @@
+[pbs_params]
+    lwe_dimension=839
+    glwe_dimension=1
+    polynomial_size=2048
+    lwe_noise_distribution={TUniformBound= 4}
+    glwe_noise_distribution={TUniformBound= 17}
+    pbs_base_log= 23
+    pbs_level= 1
+    ks_base_log= 2
+    ks_level= 7
+    message_width= 2
+    carry_width= 2
+    ciphertext_width= 64
+    opportunistic=true
+
+[ntt_params]
+    core_arch= {GF64=[5,6]}
+    min_pbs_nb= 11
+    batch_pbs_nb= 12
+    total_pbs_nb= 32
+    ct_width= 64
+    radix= 2
+    stg_nb= 11
+    prime_modulus= "GF64"
+    psi= 64
+    delta= 5
+
+[ks_params]
+    width= 21
+    lbx= 3
+    lby= 64
+    lbz= 3
+
+[pc_params]
+    ksk_pc= 16
+    ksk_bytes_w= 32
+    bsk_pc= 8
+    bsk_bytes_w= 32
+    pem_pc= 2
+    pem_bytes_w= 32
+    glwe_bytes_w= 32
+
+[regf_params]
+    reg_nb= 64
+    coef_nb= 32
+[isc_params]
+    min_iop_size= 4
+    depth= 64