* Tolerate differences in RSA private key libraries

* simplify slice comparison

- See definition in Part 3, Commands, section 15.4

* Improve TPMUSymKeyBits and TPMUSymMode

* Add Deprecated comments

* Small corrections

* Add Deprecated comments

* Implement TCP TPM protocol

This change implements the basic TCP TPM protocol, which is useful for
connecting to a running TPM simulator. It also introduces some tests
which are skipped if the --cmd_addr or --plat_addr flags aren't
provided.

A number of TCP commands aren't implemented and left for a future
change.

Tested against a recent build of the TCG TPM simulator from
https://github.com/trustedcomputinggroup/tpm

use the closeTPM helper, caught by linter

Co-authored-by: anamendes23 <anamendes@google.com>

* Fix issues raised in PR

---------

Co-authored-by: anamendes23 <anamendes@google.com>

Signed-off-by: loicsikidi <loic.sikidi@gmail.com>

* An improved check whether the "sel" argument of encodeTPMLPCRSelection() is a default PCRSelection{} object (whose fields are unset).

While encodeTPMLPCRSelection() already correctly returns tpmutil.Pack(uint32(0)) in case "sel" is a default PCRSelection{} object, we are planning to add support for TPM2_PCR_Allocate command, and then the following "sel" value may be used with TPM2_PCR_Allocate command:
[
    {Hash: tpm2.AlgSHA1, PCRs: []int{}},
    {Hash: tpm2.AlgSHA256, PCRs: []int{0,1,2,...,23}},
]"
The existing code with the "len(s.PCRs) == 0" check would produce an incorrect result with this "sel".

An example from section "22.5 TPM2_PCR_Allocate" of the spec https://trustedcomputinggroup.org/wp-content/uploads/Trusted-Platform-Module-2.0-Library-Part-3-Version-184_pub.pdf
"To change the allocation of a TPM from 24 SHA1 PCR and no SHA256 PCR to 24 SHA256 PCR and no SHA1 PCR, the pcrAllocation would have to have two selections: one for the empty SHA1 bank and one for the SHA256 bank with 24 PCR."

* tpm2: Add TPM2_PCR_Allocate support.

Some TPM chips "pad" the EK cert that is stored in the TPM to completely
fill up the NV index. Most TPM chips do NOT do this. If we try to parse
the certificate with this extra padding, the x509.ParseCertificate
function will return an "trailing bytes" error.

Before trying to parse the certificate, we need to identify how many
bytes should be stripped from bytes that were read from the TPM. To do
this, we can use the ASN.1 Unmarshal function to get the "padding"
bytes. From there, we can use the length of the padding bytes and remove
those bytes from the bytes we send to the x509.ParseCertificate function
so it will parse correctly.